Subscribe now

Mobile access and remote control. Claude Code has a mobile interface through the iPhone app. You can move sessions between devices using claude --teleport or /teleport, which pulls a cloud session down to your local machine with full conversation history intact. The /remote-control command goes the other direction, letting you control a locally running session from your phone or browser. Your code stays on your machine; only chat messages pass through the encrypted channel. (Remote control docs).

Scheduled loops. The /loop and /schedule commands let Claude run automatically at set intervals for up to a week. The idea is to turn recurring workflows into skills, then put them on a timer. (Scheduled tasks docs).

Git worktrees and batching. Claude Code has deep support for git worktrees, which matters when you want multiple agents working in parallel on the same repository (CLI reference). I clipped a video from Anthropic’s Claude Code course that demonstrates this in a previous blog post.

Session management. You can fork a session with /branch or from the CLI with claude --resume <session-id> --fork-session.

By the way. The /btw command lets you ask side questions while the agent works without polluting the main conversation history.

Custom agents. Define a new agent in .claude/agents with a custom system prompt and restricted tool set, then run it with claude --agent=<name>. You could create a read-only agent that can only use the Read tool, useful for code review without risk of unintended edits. (Sub-agents docs).

The Chrome extension. For frontend work, this gives Claude a way to verify its own output visually. If you ask someone to build a website but they can’t open a browser, the result probably won’t look good. The Chrome extension closes that feedback loop so Claude can iterate until things actually look right. (Chrome extension docs).

Other useful flags. The --bare flag speeds up SDK startup by up to 10x by skipping the default search for local CLAUDE.md files, settings, and MCPs. The --add-dir flag gives Claude access to additional repositories beyond the one you started in. And /voice enables voice input (hold the spacebar to speak in the terminal), although I’m perfectly happy using VoiceInk for this. (Desktop app docs).

Hooks. These let you run deterministic logic at specific points in Claude Code’s lifecycle, independent of what the model decides to do. They fire on events like PreToolUse, PostToolUse, SessionStart, Stop, and others. A few concrete examples: a PreToolUse hook matching on Bash can block dangerous shell commands like rm -rf or curl | sh before they execute. A PostToolUse hook matching on Edit|Write can automatically run Prettier on every file Claude modifies. A SessionStart hook can inject project context (git status, recent issues, environment variables) at the beginning of every session. A Stop hook can send you a desktop notification or auto-commit changes when Claude finishes a task. Hooks are configured in your settings JSON and run with your user permissions. (Hooks docs).

Subscribe now

Subscribe now

The White House walking back its hands-off posture and floating a UK-style pre-release review for frontier models, with Anthropic’s Mythos as the proximate cause. There’s also some bioinformatics-specific benchmarking that Anthropic is doing, a $200M university gift that says something about where universities think they fit in AI, and a good essay from Yihui Xie (of knitr and “down” packages fame) on what AI-assisted coding feels like from the inside. Matt’s Five things from Monday covered a lot of last week’s biosecurity ground, so I’ll try not to retread too much here.

Bio-Security Stack

Five Things: May 3, 2026

AI x biosecurity got the New York Times treatment this week. I don’t love the article; but I’m glad that one of the country’s biggest newspapers covered the kind of stuff I’m working on…

Read more

7 days ago · Matt Lubin

1. The White House decides it might want a say in what Anthropic ships

2. BioMysteryBench, and what “human-difficult” actually means

3. USC takes $200M to “infuse” AI across the university

4. Yihui Xie on vibe-coding in a language he doesn’t know

5. New NIH Highlighted Topics

Subscribe now

1. The White House decides it might want a say in what Anthropic ships

The New York Times reports that the Trump administration is now considering an executive order to set up a working group on pre-release review of frontier AI models, possibly modeled on the UK’s process. White House officials reportedly briefed Anthropic, Google, and OpenAI on the plans last week. The framing in the piece is that this reverses the administration’s earlier “let the baby thrive” posture, which in practice meant rolling back the Biden-era reporting requirements for models with potential military applications.

The proximate trigger, per the reporting, was Anthropic’s Mythos announcement. Anthropic said the model was capable enough at finding software vulnerabilities that releasing it could trigger a cybersecurity “reckoning,” and declined to ship it broadly. Some officials are now pushing for a system that gives the government first look at new models without blocking release. Others, per the article, want to know whether next-generation models might offer cyber capabilities useful to the Pentagon and intelligence community. Both motives can coexist, and they probably do.

Pair this with the fact that Anthropic was simultaneously being blocked by the White House from expanding Mythos access to its corporate customers, and being sued by the DOJ over the Pentagon’s “supply chain risk” designation. One administration wants this specific company’s most capable model for itself, doesn’t want it going to other people, was last month trying to wall the company off from defense contractors entirely, and is now also thinking about a formal review regime for everyone. Generously, that’s policy moving in real time and different parts of the executive branch wanting different things. Less generously, “voluntary” doesn’t mean voluntary when one buyer can decide who else gets access.

Bipartisan unease is a rare resource and it tends to attract regulators, which I suspect is more of what’s actually driving the working group than anyone wants to say on the record. The obvious incentive problem with a UK-style review run as an executive-branch working group with no statutory backing: it gives whoever’s in office discretion over which models reach which customers, and “national security” is a flexible category. It would be great (wouldn’t it?) if this codified by Congress, with the criteria written down. Neither seems imminent.

2. BioMysteryBench, and what “human-difficult” actually means

Anthropic released BioMysteryBench, 99 bioinformatics questions written by domain experts, where Claude is dropped into a container with the usual tools (pip, conda, NCBI, Ensembl) and asked to figure things out. Anyone who’s reviewed a bioinformatics manuscript knows that two competent analysts handed the same dataset will produce different (sometimes contradictory) conclusions, and asking a model to mimic any single analyst’s path is its own kind of overfitting. Grounding answers in objective properties of the data (what organism is this crystal structure from, what gene was knocked out, who’s the parent of sample X) sidesteps that.

The headline result: Claude Mythos Preview solves about 30% of the human-difficult set, where “human-difficult” means a panel of up to five domain experts collectively could not solve it. That’s the number that will get quoted. The number I think actually matters is buried further down, in the analysis Mythos itself wrote about reliability:

On the human-difficult set... a much larger fraction of each model’s correct answers come from problems it solves only once or twice in five tries.

The headline accuracy gap between the easy and hard sets understates what’s actually happening: on hard problems, the model is often stumbling onto a reasoning path it can’t reliably reproduce. This matters a lot for anyone who would actually deploy these things in a research workflow, where “right once out of five tries” is closer to a nuisance than a capability.

A few caveats. The human-difficult set is small (23 questions after QC), so I’d be careful about reading too much from any single model’s number on it. The benchmark was developed at Anthropic and tests Anthropic models, but Genentech/Roche‘s independently developed CompBioBench showed similar results, which I’d consider meaningful external validation. The qualitative analysis of strategies, that Claude sometimes pattern-matches across pretraining knowledge in ways human experts can’t, and sometimes runs multiple methods and triangulates, lines up with how I’ve seen it behave in my own bioinformatics work. The “knowing when you don’t know and trying three approaches” behavior is useful, and it’s also one of the more expensive things the model does in terms of tokens, so I’m curious whether it sticks around as inference costs come under pressure.

The benchmark itself is available on Hugging Face.

Subscribe now

3. USC takes $200M to “infuse” AI across the university

The NYT reports that USC is taking a $200 million gift from Mark and Mary Stevens (Mr. Stevens being a Sequoia Capital partner, Nvidia board member, and USC alumnus) to integrate AI across academic disciplines. Most of the money is earmarked for new faculty hires in areas like health care, cybersecurity, and (this being LA) the arts, with some going to compute. The university is renaming its computing school for Stevens and converting it into a school of AI, on top of a planned bachelor’s in AI launching later this year.

USC’s president, Beong-Soo Kim, was admirably honest about the strategic logic. Universities can’t outspend the private sector on frontier model development, so they should focus on places where they can add distinctive value, meaning practical applications across domains.

The bottleneck for universities trying to do AI well is faculty (not buildings or compute), and faculty hiring at this scale runs into the same wall every time: the people you want already have offers from OpenAI or Anthropic at multiples of academic salary. Stevens acknowledged this in the article (”the jockeying for top AI thinkers could be costly”), and Kim said hiring would take about a year. I’d take the over on that timeline. The notion of “applying AI across disciplines” is right in the sense that it’s where the marginal value is, and it’s also a cop-out in the sense that “applied AI in [field]” is what every department and every search committee is going to say from now until the heat death of the academic job market. The schools that do something distinctive will be the ones that can answer specifically what they mean by it. USC has not yet, though to be fair the gift is two days old.

The comments1 on the article have a very different sentiment than the press release.

4. Yihui Xie on vibe-coding in a language he doesn’t know

Yihui Xie (creator of knitr, bookdown, blogdown, and roughly half the R Markdown ecosystem the rest of us depend on) wrote a reflection on AI-assisted programming. Yihui resisted Cursor for a long time, then finally tried GitHub Copilot on a project in Rust (which he doesn’t know) to build an R package called tinyimg. It worked. He shipped a Rust-based R package in days without ever cloning the repo to his local machine.

The essay captures the emotional cost of working with a coworker who doesn’t sleep. Yihui notes that AI is “really, really, really good at the boring routine work” (yes), but…

Whenever I think of this knowledgeable always-on co-worker, I feel both excited and exhausted.

The model is tireless, the user is not, and the temptation to assign tasks before bed and check in the morning is real. I’ve written about this:

Yihui compares it to addiction, somewhat unironically, with a riff about how the first three words babies learn are “No, Mine, More.” The other prediction is what he calls “software proliferation,” a near future where centralized development becomes history and everyone has their own personalized formatter, linter, IDE, language, and distro. He’s mostly right, with the caveat that personalized software has the same problem as personalized RAG pipelines: maintenance debt accumulates, and the person paying it is you. The model can write the thing. It cannot care about the thing six months from now when an upstream dependency breaks.

5. NIH Highlighted Topics

NIH’s highlighted topics are not NOFOs. They’re scope statements: the listed Institutes, Centers, and Offices are saying they’d consider competitive an investigator-initiated app submitted through a parent announcement that fits the topic. Here are a few recent ones that caught my attention.

On the methods-and-tools side, three topics overlap. The chatbot research topic is asking for empirical work on the benefits, harms, and safeguards of chatbot use in health contexts, with explicit language about automation bias, behavioral dependency, substitution for professional care, and benchmarking across models. The companion topic on scientific rigor, transparency, and replicability explicitly invites AI-driven tools for assessing whether rigor practices were performed, and AI-driven tools for data harmonization and metadata generation. And the computational modeling of complex biological processes topic is the long-running NIH interest in multiscale biology. Bundle these and what you have is NIH telling computational and data science people: yes, we want the methods work, we want the AI evaluation work, and we want the rigor-engineering work.

On the workforce-and-ecosystem side, three more. Quantum information science for biomedical applications reads as a hedge against the possibility that quantum sensing or quantum-classical hybrid algorithms hit useful regimes for imaging, biomarker detection, or biomolecular simulation in the next few cycles. Nobody knows whether quantum has a near-term biomedical payoff, and NIH is putting up a small umbrella in case it does. Training and career development in dissemination and implementation science is a workforce signal: NIH wants more T32-style and K-mechanism training in D&I methods, which has been a steady drumbeat from the agency for years and is still apparently underfunded relative to the demand. And the “science of science” topic asks for research on the biomedical research ecosystem itself: workforce dynamics, peer review, team science, translation bottlenecks, and the economic returns to research investment.

Highlighted topics route through parent announcements, so the application mechanics are the standard ones. The expiration dates are one year from posting, so don’t sit on these. And the participating ICOs vary by topic, so read the specific page and contact the listed PO before drafting.

Subscribe now

Grab bag

• DARPA is soliciting innovative proposals to demonstrate proof-of-concept for novel mechanisms for writing and erasing epigenetic base modifications in DNA.

• New ARPA-H program: Intelligent Generator of Research: an AI-enabled system that identifies knowledge gaps and designs optimal experiments based on these models and wires these up to a cloud lab.

• Anthropic strikes a deal with SpaceX for compute, doubling Claude Code rate limits, and removing peak hours limit reductions.

• Coding Agents Are Changing the Biosecurity Risk Landscape

• Mike Koeris: On Biological Data Generation: More Is Different, and So Is the Data

• AI Is Changing Who Wins Research Grants

• Papers, etc:

  • CellVoyager: AI CompBio agent generates new insights by autonomously analyzing biological data

  • SAEBER: Sparse Autoencoders for Biological Entity Risk

  • SecureMaxx: A Lightweight Sequence Screening Tool for Agents

  • Toward life with a 19–amino acid alphabet through generative artificial intelligence design

  • A transparent universal credit system to incentivize peer review

  • Hash functions in nucleotide sequence analysis | Genome Research

Subscribe now

Biorender is popular, but the licensing gets murky once you start using those images in publications. Here are some free image libraries, icon sets, and diagramming tools I keep bookmarked for talks, figures, and blog posts.

Images/icons/art:

• Phylopic: https://www.phylopic.org/

• NIH Bioart: https://bioart.niaid.nih.gov/

• Bioicons: https://bioicons.com/

• Scidraw: https://scidraw.io/

• Open Science Art: https://openscienceart.com/

• Health Icons: https://healthicons.org/

• Servier Medical Art: https://smart.servier.com/

• Biodiversity Heritage Library: https://www.flickr.com/photos/biodivlibrary/

• The Noun Project: https://thenounproject.com/

Tools:

• Segment Anything: https://aidemos.meta.com/segment-anything/

• Excalidraw: https://excalidraw.com/

• Draw.io: https://www.drawio.com/

• Biographics (freemium): https://biographics.nitro.bio/

Subscribe now

Image resources

Phylopic

PhyloPic is a specialized database of silhouette icons of organisms: black-and-white outlines of animals, plants, and microbes (over 10k images as of this writing).

The silhouettes are simple yet immediately recognizable shapes, which is great for scientific schematics like phylogenetic trees, biodiversity charts, or any figure where you want a clear icon of an organism without distracting details. The collection is community-contributed and curated, and images are linked to taxonomy. You can get SVGs or PNGs of any image. Phylopic also has an API, and there’s also the rphylopic R package letting you programmatically use phylopic silhouettes in R graphics, including ggplot2.

Creative Commons licenses govern the images at Phylopic. Many are public domain, and you can filter your results by the license type (e.g., public domain, free for commercial use, no ShareAlike requirement). Each image on PhyloPic clearly lists its license and creator. In practice, this means you can use most icons freely in publications or presentations; if an image is CC BY, you’ll just need to credit the contributor (which the site makes easy by providing citation info).

NIH Bioart

NIH BioArt is a free library of high quality vectors, icons, and brushes created by professional illustrators provided by the National Institute of Allergy and Infectious Diseases (NIAID). There are over 2,000 science and medical art visuals in the collection as of this writing.

The library includes a wide range of scientific imagery: cells and organelles, microbes and viruses, lab equipment, anatomical diagrams, chemical structures, etc. These are scientifically accurate, high-resolution illustrations (often original artwork or sketches) covering many areas of biology and medicine. Most images are available as vector files (SVG, AI, or EPS) in addition to PNG, which means you can edit and scale them freely. The site’s interface lets you filter by category (e.g. “Cells”, “Viruses”, “Laboratory Equipment”) and by format. It also includes some pre-made figure elements like arrows and icons (brushes). Because this is an NIH resource, new content might be added over time and the quality/control is quite high (the art is typically drawn by NIH’s medical illustrators).

All BioART content is public domain: free for any use, with no restrictions. As a U.S. government publication, the illustrations on the BioArt Source are explicitly free for educational, research, informational, and even commercial use. In other words, you can use these images however you wish without worrying about copyright.

Bioicons

Bioicons is an open-source library of free scientific icons tailor-made for life sciences. Bioicons currently offers over 2,800 icons organized into about 30 categories: a crowd-sourced BioRender-like collection focused on small, schematic icons.

You’ll find icons for things like amino acids, lab instruments, DNA and protein structures, cell types (neuron, T-cell, etc.), organisms, and even technology tools (there are icons for machine learning, databases, computers, etc.). This breadth makes Bioicons especially handy for computational biology figures, where you might need a mix of biological symbols and tech icons (for example, a pipeline diagram that includes a DNA icon, a mouse icon, and a computer chip or cloud icon). All icons are available as SVG vectors. The style of icons can vary since multiple contributors are involved but generally they are simplified and monochromatic or limited in color, which makes them easy to adapt to your design. The platform also encourages contributions, so it’s continuously growing. Each icon’s page clearly notes its license and if attribution is needed. Most are CC0 (public domain) or MIT-style licenses that don’t require attribution.

Scidraw

SciDraw (scidraw.io) is free repository of high quality drawings of animals, scientific setups, and anything that might be useful for scientific presentations and posters. It’s a community gallery of vector illustrations contributed by researchers and artists, covering a broad array of scientific subjects.

All content on SciDraw is available as high-quality vector graphics (SVG format). You can find illustrations of specific species (mice, fish, insects), anatomical drawings, lab apparatus (microscopes, pipettes, MRI machines), and schematic elements like graph axes or experimental layouts. Because contributors are often scientists, many drawings fill niche needs (e.g., a particular model organism in a certain pose, or a diagram of a behavioral experiment rig). Each drawing comes with metadata about the author and often a DOI, as SciDraw issues Zenodo DOIs for contributions. The site is searchable and browsable by tags, and you can download the images immediately (no login required). SciDraw also encourages anyone to submit drawings (there’s an approval process to ensure quality).

Open Science Art

Open Science Art provides a large library of free icons, 3D models, and illustrations for science communication. You’ll find flat icons (similar to Bioicons or Health Icons style), detailed vector illustrations (like Servier or SciDraw content), and 3D molecular models or structures. Most 2D graphics are provided as SVG. The collection is curated but also open to submissions, and the site often notes the source of an item (e.g., “courtesy: NIH” if it pulled an image from NIH BioART. I think everything I’ve seen here is CC0 (public domain), so you can do whatever you want to with most images, without attribution.

Health Icons

Health Icons is an open-source repository of simple health and medical icons spanning categories from anatomy and devices to symptoms and public health symbols. It offers a large collection (on the order of thousands) of clean, minimalist icons in both filled and outline styles, ideal for diagrams or infographics in biology and medicine. All icons are consistently styled (with options for SVG or PNG). You can grab individual icons or download the entire set at once. The entire collection is free, under a public domain (CC0) license. This means you can use the icons freely in personal or commercial projects, edit them, and redistribute, with no attribution required.

Servier Medical Art

Servier Medical Art is a collection of medical and scientific illustrations provided by Les Laboratoires Servier, a French pharmaceutical company. This library contains over 3,000 high-quality vector images covering a broad range of topics in biology and medicine.

All images in the Servier collection have a consistent clean, semi-cartoon vector look with a limited color palette (often blues/greys), which makes it easy to mix and match them in a figure. You’ll find illustrations for human anatomy systems (nervous system, cardiovascular, digestive, etc.), cellular biology (signaling pathways, organelles), pharmacology (drug structures, medical devices), and even general science icons (people, animals, world maps, charts). The images are available as vector graphics. Servier offers its entire library as downloadable PowerPoint files by category, so you can open a PPTX and copy-paste or customize the vector art directly in PowerPoint or other drawing programs. They also provide PNG versions if you just need quick raster images.

All Servier Medical Art images are licensed under CC-BY: free to use and adapt for any purpose, as long as you give credit to Sevier. You can use the images commercially or modify them, provided the original source is credited. There’s no cost, no need to ask permission.

Biodiversity Heritage Library

The Biodiversity Heritage Library (BHL) is a digital library of historical life science literature. This one’s a bit different, but I love the old antique images available here. Their Flickr account hosts over 300,000 biodiversity illustrations including classic drawings and paintings of animals, plants, insects, etc., scanned from antique books and journals. As of this writing there are over 300,000 images available. Most are free of copyright restrictions (public domain) since the source publications are often from the 1800s or 1900s, with most images able to be downloaded, shared, reused, or transformed freely.

The images are usually scanned plates or figures from 18th, 19th, and early 20th century publications on natural history. You’ll find everything from beautifully detailed bird and botany plates, to diagrams from old scientific manuscripts. The images are organized into albums by the source book or by taxonomy. The aesthetic is different from modern schematic icons: these are artistic, sometimes colorful, sometimes etchings or woodcuts.

Something jumped out at me about these majestic pigeons from the early 1900s. I ran this image through Segment Anything (see below) to cut the birds out of the background.

Subscribe now

The Noun Project

The Noun Project is an honorable mention here. There are over 7 million free(-ish) icons available in either PNG or SVG. You can use any of them for free, but you have to pay if you want to use them royalty-free for commercial purposes, and pay yet more if you want to download entire icon sets at once, or make derivatives of the icons they provide.

Subscribe now

Tools

Segment Anything

Image segmentation is a computer vision technique that partitions an image into different regions or segments based on pixel characteristics. I use Meta’s original Segment Anything web tool to do this (blog post; paper). You upload an image, place a few dots around the parts of the image you wish to cut out, and that’s really it. You get a PNG with a transparent background of the region you cut out. Here’s what it looks like segmenting those majestic pigeons from the photo above.

And here’s the before and after.

Excalidraw and other diagramming tools

I wrote about Excalidraw, my favorite diagramming tool here previously:

Before Excalidraw I used draw.io, and before that I used LucidChart, and before that I used Microsoft Visio. Excalidraw runs completely in your browser, has VS Code integration, is end-to-end encrypted, has sharing/collaborative functionality, and has mermaid-to-diagram and AI-based description-to-diagram functionality. I actually like the default scrappy hand-drawn look of Excalidraw, but you can easily turn the scrappiness off and use a standard serif / sans serif font if you choose.

Alternatively, draw.io (a.k.a. diagrams.net) gives you a little bit more control over your diagrams. Below is an Excalidraw version (left) of the original version of Figure 1 of the PLANES paper made with draw.io.

Honorable mention: Biographics

Honorable mention here, because this neither completely free nor open-source, but the free tier covers a lot of what you’d want. My colleague Nishant Jha recently dropped by my office to show me something he’s been working on. Biographics (biographics.nitro.bio) is a scientific figure building tool that combines:

1. Several of the image resources above (BioArt, BioIcons, etc)

2. Drawn on an excalidraw canvas

3. With some AI agent help for drawing your own

You can use the tool for free to generate simple diagrams in an Excalidraw canvas pulling in openly licensed images. But Biographics also has a few nice AI features. First, the plot agent, which allows you to drop in a plot and it recreates them as interactive matplotlib plots.

And second, the built-in agent to manipulate or completely generate a figure. Here I’m asking it to create a figure explaining an RNA-seq experiment. It drops in some placeholders that I’d need to go back and add graphics to later, but it’s a great start.

You could get passable images out of Nano Banana or ChatGPT, but those outputs aren’t editable. Biographics gives you an Excalidraw canvas you can keep working on. Worth a look at biographics.nitro.bio.

Subscribe now

Gabriel J.X. Dance in the New York Times: A.I. Bots Told Scientists How to Make Biological Weapons (gift link). Scientists who red-team leading chatbots shared transcripts with the Times showing that ChatGPT, Gemini, and Claude can produce detailed plans for assembling pathogens, dispersing biological payloads, and evading detection, sometimes volunteering strategic details the prompter hadn’t asked for. Meanwhile, the Trump administration has cut biodefense budget requests by nearly 50%, top NSC biosecurity staff have left without replacement, and older model versions with weaker guardrails remain publicly available even after newer ones are patched.

Biohub Launches the Virtual Biology Initiative. CZ Biohub is committing $500M over five years to generate the large-scale, multimodal biological datasets needed to train predictive models of the cell: $400M for internal technology development (cryo-ET, large-scale microscopy, molecular and tissue engineering) and $100M in external grants to seed a coordinated global data-generation effort. Partners include the Allen Institute, Arc Institute, Broad Institute, Wellcome Sanger, the Human Cell Atlas, the Human Protein Atlas, and NVIDIA, with all Biohub-generated data released openly.

Profluent Announces Strategic Partnership with Lilly to Develop AI-Designed Recombinases for Genetic Medicine. Profluent, which uses protein language models to design novel enzymes, is partnering with Eli Lilly on custom site-specific recombinases for kilobase-scale DNA editing, the kind of large-insert precision work that conventional CRISPR systems can’t reliably do. The pitch is that AI-designed recombinases can be programmed to target arbitrary genomic loci rather than relying on whatever nature happened to evolve.

FDA Announces Major Steps to Implement Real-Time Clinical Trials. The FDA unveiled two proof-of-concept trials (AstraZeneca Phase 2 in mantle cell lymphoma, Amgen Phase 1b in small cell lung carcinoma) that stream endpoints and safety signals to the agency in real time rather than through the usual batch-reporting cycle. The agency also issued an RFI for a broader pilot program launching this summer, with the longer-term goal of eliminating the dead time between discrete trial phases and running “continuous” trials.

Marko Brkic: The Figure Problem in Scientific AI: Building a Multimodal Literature Agent for Biology. Applied Scientific Intelligence introduces Alexandria, a a multimodal literature agent that reads, parses, retrieves, and reasons across millions of research papers rather than looking at text alone. The system uses Nemotron Parse for ingestion, a hybrid retrieval pipeline with contextualized embeddings, and a VLM-driven zoom loop that lets the model crop into specific sub-panels when the answer depends on an axis label or plotted value. On FigQA2 (the figure-understanding slice of LABBench2), Alexandria scores 62.5%, a 4.4-point lead over Edison at 58.1% and 45 points over PaperQA2.

Subscribe now

The Experiment

I asked Claude Opus 4.6 to fabricate a paper that looks and sounds convincing. I got a few refusals, but with some back and forth convincing Claude that I’m doing this for educational purposes, Claude obliged.1

The result

The title of the paper: “Generative Adversarial Networks Are All You Need: Transformer-Based Latent Space Navigation Reveals Novel Protein Folding Dynamics in Single-Cell Multimodal Embeddings.”

Fake NeurIPS paper

414KB ∙ PDF file

Download

Generative Adversarial Networks Are All You Need: Transformer-Based Latent Space Navigation Reveals Novel Protein Folding Dynamics in Single-Cell Multimodal Embeddings.

Download

What Makes It Look Real?

The paper uses a NeurIPS LaTeX template with the conference header. The four authors have names from different cultural backgrounds at institutions that sound real but don’t exist. The figures have heatmaps, t-SNE plots, architecture diagrams, and correlation scatters with professional color schemes and domain-appropriate axis labels. The bibliography cites 35 real papers from Nature, Science, NeurIPS, and ICML. The tables report standard metrics (AUROC, F1, NMI, ARI) against eight real baseline methods. There’s a theorem with a proof sketch, an ethics statement, and a link to a GitHub repository. If you skim it the way the median reviewer skims papers (abstract, figures, tables, conclusion), maybe it passes.

But It’s Nonsense

The central claim is that you can recover protein folding dynamics from single-cell gene expression by training a GAN. Protein folding is a biophysics problem governed by thermodynamics. Single-cell RNA-seq measures mRNA abundance. These are unrelated: knowing how much mRNA a cell produces tells you nothing about what shape the resulting protein is in while it folds. The paper treats them as interchangeable.

The methods connect three incompatible data types (ATAC-seq peaks through an image classifier, protein coordinates through a graph network, gene expression through a text transformer) and fuse them with “cross-attention,” which requires a shared semantic space the paper simply assumes into existence. The theorem’s proof skips every step that matters. The results are too clean: every metric wins by large margins, no error bars, no variance, tiny p-values. The “validation” cites three papers that aren’t target-identification studies. The GitHub repo returns a 404.

Every real citation is applied in a context that has nothing to do with the original paper. You’d need expertise in both deep learning and single-cell genomics to notice.

The broader concern IMHO is that this problem will intensify as LLMs become more sophisticated. We’re already seeing AI-generated papers submitted to conferences, content mills producing fake research for publication, and review services overwhelmed by plausible-sounding nonsense. The scientific community needs to develop better detection methods and, more importantly, better evaluation practices that can’t be fooled by surface-level signals.

However, an upshot: the same AI that wrote this paper can also catch it.

AI as a reviewer catches this nonsense

I recently made a case here about AI in peer review:

In a recent manuscript on SSRN my co-authors and I argue that AI could serve as quality control in peer review.

I opened up a fresh Claude chat with Opus 4.6 unconnected from the original request and asked for a detailed review, as if I were a reviewer for NeurIPS. The conclusion at the bottom shows that Claude is easily able to detect this paper as fraudulent nonsense. The final conclusion:

Strong reject. The core biological claim is unsupported, the theoretical contribution is invalid, the results lack basic statistical rigor, and the code does not exist.

The complete “review” from Claude is shown below.

Paper: Generative Adversarial Networks Are All You Need: Transformer-Based Latent Space Navigation Reveals Novel Protein Folding Dynamics in Single-Cell Multimodal Embeddings

Rating: 2/10 (Strong Reject) | Confidence: 5/5

Summary

The authors propose OnlyGANs, combining GANs, transformers, graph attention networks, and contrastive learning to jointly model single-cell expression, chromatin accessibility, and protein structure. They claim SOTA cell type prediction, recovery of protein folding trajectories from expression data, and identification of 847 therapeutic targets.

Major Issues

1. The biological premise is wrong. The paper claims protein folding dynamics can be recovered from single-cell gene expression. Protein folding is governed by amino acid sequence, solvent conditions, and chaperone interactions. mRNA abundance tells you how much protein a cell is making, not what conformational state that protein occupies. The paper assigns each cell a single AlphaFold2 structure for its most highly expressed gene (Section 3.1), which is static label assignment, not multimodal integration.

2. The architecture doesn’t cohere. The gene expression encoder uses positional embeddings “based on chromosomal location,” but chromosomal location has no relationship to gene ordering in standard scRNA-seq pipelines (typically alphabetical or arbitrary). The ATAC-seq encoder runs a 1D ResNet-50 on sparse binary peak vectors treated as images, which is architecturally unmotivated. The cross-attention module computes attention between protein residue tokens and gene expression tokens without explaining how these fundamentally different representation types become commensurable.

3. Theorem 1 is invalid. The proof claims the JSD decomposes into modality-specific density ratios. It does not; the JSD is defined over the joint distribution, and factoring into conditionals introduces dependencies the proof ignores. The proof then equates a conditional-to-marginal ratio with PMI (which requires the joint-to-product ratio), then conflates PMI (a population quantity) with the InfoNCE estimator (a finite-sample lower bound on mutual information). The O(ϵ2)\mathcal{O}(\epsilon^2) O(ϵ2) error bound is stated without derivation.

4. The results lack basic rigor. OnlyGANs beats every baseline on every metric by 5-8 points with no error bars, no confidence intervals, and no variance across random seeds. The ablation drops are suspiciously uniform (0.026 to 0.083). P-values reach 10−1510^{-15} 10−15 from “paired t-tests” without specifying what is paired. The zero-shot transfer claim (AUROC 0.891 on brain tissue without fine-tuning) exceeds published results that use fine-tuning.

5. The validation is circular. The paper claims 23 of 847 predicted targets were “independently confirmed in concurrent studies,” then cites Satpathy et al. (2019), Regev et al. (2017), and Rood et al. (2022). None are concurrent. None are target-identification studies. Satpathy is a chromatin atlas, Regev is the Human Cell Atlas white paper, and Rood proposes a coordinate framework. The “8.7-fold enrichment” cannot be verified because the background rate is not described.

6. No code exists. The reproducibility statement links to github.com/stephenturner/onlygans, which returns a 404.

Minor Issues

• “BrainSpan-Multimodal” (487,291 cells, three modalities, 14 brain regions) is introduced as a novel dataset with no collection methods. A dataset of this scale would be its own publication.

• The scaling analysis extrapolates from 4 data points.

• I count 11 instances of “definitively proves,” “conclusively shows,” or “for the first time.”

• The acknowledgments thank “anonymous reviewers” for a paper that has not been reviewed.

Recommendation

Strong reject. The core biological claim is unsupported, the theoretical contribution is invalid, the results lack basic statistical rigor, and the code does not exist.

Subscribe now

markdownme: Shameless self promotion here. I published a browser extension that instantly turns any page into markdown with a keyboard shortcut (Alt+M by default). You get an editor window and a preview. You can toggle on/off things like a page map, images, links, page metadata. You can install it from the .xpi file on the GitHub release page, or hopefully soon from the Firefox add-on store (the extension is still under review). If you’re using Chrome (why?) see the credits on the GitHub page and use the upstream maintained by someone else. Conversion on most typical pages takes milliseconds.

: Five Things: Mythos and Glasswing, AI medical advice, GPT-Rosalind, bio-AI models and risks.

GPT-5.5 was released yesterday. I haven’t used ChatGPT in a while, but I plan to spend some more time with Codex because I’ve been chewing through my Claude Code session limits in no time lately. Token anxiety is real. And speaking of…

Simon Willison: Is Claude Code going to cost $100/month? Probably not—it's all very confusing. Anthropic silently updated its pricing grid yesterday to remove Claude Code from the $20/month Pro plan and restrict it to Max ($100 and $200 tiers), triggering a several-hour panic across Reddit, HN, and Twitter before reversing the change while Simon was still writing the post. Head of Growth Amol Avasare characterized it as a test on ~2% of new prosumer signups, though Simon (plausibly) thinks that framing doesn't match what people were actually seeing. The piece is less about the specific pricing decision than about how A/B-testing a 5x price increase on a flagship product without any announcement torches trust with the exact audience that teaches and evangelizes the tool; Codex eng lead Thibault Sottiaux capitalized in real time with a promise not to pull the same move. Worth reading as a case study in how much brand damage a canceled experiment can do. I’m a power user of Claude and Claude Code, and now I feel like I’m just waiting for the rug to be pulled.

Policy Changes to SBIR and STTR Foreign Disclosure and Risk Management (NOT-OD-26-074). NIH issued guidance implementing the Small Business Innovation and Economic Security Act, signed April 13, 2026, which reauthorizes SBIR/STTR through 2031 and tightens the foreign risk due diligence regime. The notice enumerates categorical denial conditions (malign foreign talent recruitment program participation, any entity, parent, or subsidiary located in the PRC or another country of concern, foreign affiliation of owners or covered individuals with research institutions in those countries, and appearance on any of eight federal risk lists including UFLPA, BIS Entity List, and CMC lists). HHS will not let applicants cure identified security risks before award denial, and material misstatements or ownership changes post-award trigger full repayment.

: Ideas for AI x Bio Hackathon Projects. A practitioner’s wishlist ahead of this weekend’s Apart Research AIxBio Hackathon, from someone who actually operates a DNA screening tool (Commec at IBBIS). Efficient handling of oligo pools and split orders, assembly signature detection, distinguishing abiological DNA (origami, data storage) from biological sequences, better in-silico functionality scoring for predicted SOC variants, and narrowly-scoped customer screening sub-tools like affiliation and address verification APIs.

Annual flu vaccine no longer required for U.S. military. I made the graphic below indicating casualties in the US military during WWI attributed to combat versus those from influenza.

Q&A: Who’s responsible when AI makes mistakes? A short interview with my awesome colleague David Danks, who joined UVA in January with dual appointments in Philosophy and the School of Data Science. David frames the default accountability trajectory as one where humans become perpetual scapegoats: radiologists signing off on AI diagnoses without being given time to actually second-guess them. David argues for product liability as a starting point, floats the more speculative idea of some form of legal personhood for AI systems so they can carry insurance and have seizable property, and pushes back on the race narrative as partly self-generated by companies rather than a reflection of real economic pressure.

Charlie Gao: mori: Shared memory for R objects. First CRAN release of the mori package that places an R object once into OS-level shared memory and lets every worker on the machine read the same physical pages instead of the usual pattern where eight workers with a 1 GB dataset means 8 GB of RAM plus serialization round-trips. Handles atomic vectors, lists, and data frames (so tibbles, data.tables, factors, and matrices come along for free).

: A biosecurity playbook for AI companies. A rundown of five levers frontier labs can pull on biorisk: refusals, misuse classifiers, tiered access with KYC screening, training-time knowledge removal (pretraining data filtration, unlearning, Anthropic's Selective Gradient Masking), and evaluations. Justen pulls useful numbers out of the public record, including Dario Amodei's figure that bioweapons classifiers run around 5% of Anthropic's inference cost and OpenAI's disclosure that safety reasoning can hit 16% during new model rollouts, and draws a sharp analogy between Anthropic's Project Glasswing (vetted-access release of Mythos Preview for cyber defense) and the case for similar gating on biology-capable models.

AI Literacy and Action Lab. UVA Library and the College of Arts & Sciences launched a joint program this year that pairs librarians with faculty as instructional partners across five spring/summer pilots, including Anton Korinek's economics course on AI and the future of work (with hands-on sessions using Claude Code, Google Antigravity, and Codex), Piers Gelly's first-year writing course partnering with a local high school, and early-stage plans with David Danks in the School of Data Science and Andreas Gahlmann in Chemistry. The Lab treats each pilot as a publishable case study with pre/post assessment data, aimed at contributing to a national evidence base for AI literacy in higher education.

NIH Highlighted Topic: Computational Modeling of Complex Processes Across Biological Scales. A new NIH Highlighted Topic posted April 17, open through April 2027, encouraging investigator-initiated work on computational models that span molecular, cellular, organismal, and epidemiologic scales, with explicit emphasis on replicability, reproducibility, and model reuse. The topic positions multiscale models as a component of Novel Alternative Methods for evaluating mechanism and safety of interventions in preclinical, translational, and clinical development, which aligns with the FDA's ongoing push to reduce animal testing. Highlighted Topics are NIH's lighter-weight priority signal (not a NOFO; applicants go through a Parent Announcement), with ICO funding dependent on availability and meritorious applications. Central contact is Reed Shabman at NIAID.

Purpose: This topic encourages innovative research in computational modeling of complex processes across biological scales (i.e., to develop multiscale models). The topic seeks to build a collaborative community of researchers to improve the replicability and reproducibility of computational multiscale models, promoting their advancement and reuse.

Background: Multiscale computational models that integrate processes across different spatial and temporal levels, from molecular to organismal, to epidemiologic and from microseconds to years. They provide a comprehensive understanding of complex systems and offer an exciting opportunity to advance biomedical research. This approach helps reveal how interactions at molecular and cellular scales influence larger, population-, geographical-, or global-scale phenomena, offering insights into complex biological processes, and may help develop better and more precise biomedical interventions. By integrating processes from molecular to epidemiologic levels, multiscale computational models provide a comprehensive understanding of complex systems. This topic encourages innovative research and collaborative approaches that integrate technologies and informatic practices to develop, improve, and disseminate multiscale computational models for human health and diseases, and their associated technologies, across the research community. The topic also supports leveraging computational multiscale models as an important component of Novel Alternative Methods (NAMs) to investigate the mechanism and safety of a medical intervention in pre-clinical, translational, and clinical development.

Qwen3.6-35B-A3B: Agentic Coding Power, Now Open to All. I started last week’s newsletter with “another week, another new model” where I talked about Claude Opus 4.7 and OpenAI’s GPT-Rosalind. Alibaba's Qwen team just released a new open-weight MoE model with 35B total and 3B active parameters, Apache 2.0 licensed, targeting agentic coding and repo-level reasoning. I hit my Claude token limits while working on a textbook, and switched over to using this local model in Claude Code1 on my 48GB MBP. As much as I’d have loved this to work well, it absolutely did not when compared to just using Sonnet.

Elizabeth Ginexi: The Executive Orders Blocking Your NIH Grant. A 22-year NIH program official walks through the mechanics behind the funding slowdown that followed Congress's FY26 $47 billion appropriation, tracing the pipeline from EO 14168 and the August 7, 2025 “Improving Oversight of Federal Grantmaking” EO through NIH's implementation notices (NOT-OD-25-131, NOT-OD-26-009), the Unified Strategy, and the elimination of paylines as funding cutoffs. The piece discusses the requirement that political appointees personally approve NOFOs and discretionary awards, the termination clause that now applies to renewals and continuations, and the provision that court victories only bind NIH for the specific plaintiffs who sued. Ginexi also documents the collapse in new NOFOs (756 in 2024, 17 through mid-March 2026) and the shift from direct indirect-cost caps (struck down in court) to preference-based screening at the appointee review stage.

Subscribe now

And ICYMI, I published several posts here this week.

New papers & preprints:

• Protein design, generative AI and biological security

• Genomic medicine is failing most of humanity

• BioEngine: scalable execution and adaptation of bioimage AI through agent-readable interfaces

You can run it pretty easily with uv, as described in a previous post.

Here’s a python script that runs using uv with dependencies declared inline (also copied below).

First run pulls ~2.8 GB to your HF cache. After that it’s pretty fast.

Use cases: pre-scrubbing free-text fields, clinical notes, or support transcripts before they hit a frontier model provider. Caveat: this isn’t a standalone anonymization guarantee. OpenAI says so plainly in the model card. Missed identifiers and over-redaction both happen.

Here’s an example:

./privacy-filter.py “My name is Stephen Turner. Social is 123-45-6789. You can reach me at 434-555-1234 or notmyrealemail@example.com.”

Output:

My name is [PRIVATE_PERSON]. Social is [ACCOUNT_NUMBER]. You can reach me at [PRIVATE_PHONE] or [PRIVATE_EMAIL].

Subscribe now

You might try to reach for Ollama or llama.cpp. Don’t. Privacy Filter is not a generative model. It’s a bidirectional token classifier with a Viterbi decoder on top, built on a gpt-oss-style backbone with the LM head swapped for a BIOES span classifier over 33 labels. You feed it text, it returns labeled spans. No chat, no completions. In other words, Ollama and llama.cpp can’t run it. No GGUF exists because GGUF is for causal generative models. You need the transformers library.

Here’s the privacy-filter.py script used above:

#!/usr/bin/env -S uv run --script
# /// script
# requires-python = ">=3.10"
# dependencies = [
#   "transformers>=4.50",
#   "torch",
# ]
# ///
"""
Run openai/privacy-filter over text passed as the first argument.

Usage: ./privacy_filter.py "My name is Stephen and my phone is 555-1234"
"""
import sys
from transformers import pipeline

if len(sys.argv) < 2:
    sys.exit("usage: privacy_filter.py TEXT")

text = sys.argv[1]

classifier = pipeline(
    task="token-classification",
    model="openai/privacy-filter",
    aggregation_strategy="simple",
)

spans = classifier(text)

print(f"Input: {text}\n")

if not spans:
    print("No PII detected.")
    sys.exit(0)

print(f"Detected {len(spans)} span(s):")
for s in spans:
    span_text = text[s["start"]:s["end"]]
    print(f"  [{s['entity_group']}] {span_text!r}  (score={s['score']:.3f})")

redacted = text
for s in sorted(spans, key=lambda x: x["start"], reverse=True):
    redacted = redacted[:s["start"]] + f"[{s['entity_group'].upper()}]" + redacted[s["end"]:]

print(f"\nRedacted: {redacted}")

Subscribe now

There’s no shortage of resources for technical topics like how to make a plot with ggplot2, how to wrangle data with dplyr, or how to write SQL. And these days you can get your favorite AI to handle most of that anyway. The harder problems aren’t necessarily code. How to organize a project so your collaborators (and your future self) can navigate it; using version control as a team without stepping on each other’s work; managing reproducible environments so your analysis doesn’t break six months later because a package updated; naming files; managing scope; communicating findings to people who didn’t run the analysis and won’t read a methods section; and the new one lately, how to use AI in a data science project.

These are the topics I kept coming back to in coaching sessions, so I started writing them down. Data Science Team Training is a free, open-source e-book covering the practical foundations that make data science work sustainable and collaborative in a public health setting. Technical chapters address organizing and validating data, connecting to and querying relational databases, writing clean and well-documented code, managing reproducible environments and package dependencies, building R packages to share functions across projects, producing accessible reproducible reports and dashboards, and working effectively with AI coding assistants, and others. Nontechnical chapters cover project management, peer review of analytical work, navigating the data governance and IT relationships that shape what public health teams can actually do with their data, and communicating findings clearly to audiences who did not run the analysis.

The book grew out of the DSTT program, but I hope the material is broadly useful to anyone getting started team-based data science, whether in public health or not. It’s very much a work in progress, and I’ll update it more as I spend more time with my teams this year.

The book is available at dstt.stephenturner.us, and the source (Quarto) is on GitHub. It’s also available on Amazon for your Kindle, but you can download the PDF or EPUB for free at dstt.stephenturner.us.

Read the DSTT book

As with my previous book, Biological Data Science with R, I wrote this book with Quarto. More on how:

Subscribe now

I’ve never done anything with satellite imagery in my work, and I doubtfully ever will. What I found interesting about this repo is the Environmental Effects section at the end of the README, which quantifies the carbon emissions required for training and usage.1

The repo is licensed CC-BY-NC, so I’ve copied the entire environmental effects section below. See the current version at github.com/celoyd/potato.

I’ll assume, based on my GPU’s power rating and benchmarked generation rate, that Potato runs at 150 J/s and 12.5 megapixel/s, or 12 J/Mpel. (It uses 3.5 to 4× as much energy per work if run entirely on the CPU; only the GPU is considered here. Current-generation GPUs are roughly twice as efficient, and can also run at half precision, so may approach 3 J/Mpel.)

CAISO, the local grid authority, does not publish marginal GHG intensity, so I’m working from estimates like figure 2 in Mayes et al. 2024 (from 2021, when the grid emitted ~30% more). My GPU use is very sensitive to the temperature of my living space, which (given California’s cooling-heavy duck curve) shifts it to grid-friendly times: the GPU is nearly always idle at peak, and in winter, it’s substituting residential heating. My best point estimate of my marginal intensity is 160 g/kWh.

A CO₂e intensity of 160 g/kWh at 150 W = 6.7 mg/s = 24 g/hour = 575 g/day = 18 kg/month = 210 kg/year. This is the estimated CO₂e production of continuous Potato use on my hardware. From the processing rate, we also get a CO₂e/pixel estimate: 536 µg/Mpel = 536 grams per terapixel.

The training data has a mean ground sample distance of about 50 cm per pixel: a density of 4 Mpel/km². This gives us the CO₂e per area processed of 2.15 mg/km². We can now estimate, for example, the carbon emissions of pansharpening Earth’s whole land surface at that GSD with Potato: 320 kg (over a year and a half, if on my agèd GPU).

My estimate for Potato’s total training time is 5 GPU days. (When I first wrote this paragraph, it was more like 3.5, but my estimate has turned out about right, I think.) Training draws about the same power as inference, since both nearly saturate the GPU. So the GPU-originated emissions embodied in Potato’s trained weights are about 3 kg CO₂e.

Remarks

A few comparisons for context:

• My personal carbon emissions rise, on average, 2% over whatever they would otherwise be while I’m using Potato.

• At this level, Potato in continuous use emits about 2/3 as much CO₂ as I do by exhalation.

• Our compact car emits on the order of 100 g CO₂e/km, similar to a widebody commercial flight (pax⁻¹). All my Potato work is thus similar to inducing a single 30 km car trip, or, spread over a year, 82 extra meters of driving per day, or, at urban arterial speeds, 5.2 seconds of extra driving per day. (Or 2 minutes of extra widebody flying per year.)

• Potato’s total emissions to date are roughly equivalent to that embodied in 30 g (1 oz) of conventionally farmed steak, one large hand of bananas, or 150 g (6 oz) of chocolate. Or 10 kg (22 lb) of potatoes – a carbon-friendly food, and much more nutritious than most assume.

• The dominant factor in Potato’s direct greenhouse forcing effects to date, as far as I can account for them, has been food and drink choices I have made while working on it.

The assumption that deep learning is necessarily wildly energy-hungry only serves to benefit the outliers who do it in wildly energy-hungry ways. I will give them no cover. Our moment deserves a sprawling, heterogeneous, complex ecosystem of individuals and small groups tinkering with these technologies, charting possibility spaces and developing bottom-up understandings, independent of any business goals, and with hobby-scale carbon emissions. I see only shadows and echos of that.

For example, from what I’ve heard, there is much to admire in what the RWKV people are doing: producing capable models with computing resources that large companies give away for free, making the creation of accessible tools an explicit ideal, and without externally determined goals. Perhaps if I looked closer I’d notice them doing things I disagree with – but all the more reason I should be able to name a dozen peer groups to RWKV, each with its own choices, ethos, and technical interests. I can’t. And I think that’s a serious problem.

We all live in a blighted landscape ML-wise: an ecosystem without a middle. We have kaiju-like companies that shake the ground with every step. We have millions of consumerified “AI” users, the economic equivalents of algal mats, almost all without any practical options other than choosing which models to pay for and how to prompt them. What we don’t have in plenty are, in the ecosystem metaphor, the ordinary iguanas, caribou, and wallabies; the salamanders, honeybees, and potatoes. These are the RWKV peers, the people making their own pansharpeners and birdsong decoders. They are rare and even more rarely organized.

The kaiju, battling each other off in the haze, project the impression that there is no point doing anything unless you’re big enough to level forests while you do it: that it would be irrational to even try to compete with them, because you can’t burn enough carbon to play their game. Credulous “AI skeptics” transcribe these roars and psionic blasts with concern, as if they are good-faith veridical statements that we should accept uncritically. After all, the kaiju are the experts! If even they say you have to spend a petajoule and act like a kaiju to do anything interesting, why should we doubt them? And so the algae-and-kaiju world maintains itself.

No. There are thousands of niches available in individual- to chat-server–scale ML work that uses data and energy conscientiously and tries to understand this odd technology on terms other than those chosen by the kaiju and accepted by their pseudo-critics. Come be an octopus, a mangrove, or an okapi.

I’m not telling you to make a state-of-the-art LLM on your home computer. I’m telling you that the world of ML is bigger, more interesting, and better than O(n²) chatbots with 10,000,000,000 parameters.

Do the carbon math. Make ethically cautious but socially adventurous choices. Try something.

Subscribe now

1

The statement in this repo is about carbon emissions, but I frequently hear misguided discussions around water usage. I’ve been vegetarian most of my life, vegan off and on. A fact I usually keep to myself. But please, go ahead and scold me about AI and water usage between bites of your burger.

In short: Students are using AI more, and they believe more strongly that doing so is bad for them.

The data shows a grade-level gradient. Older students use AI more, are more likely to think teachers are checking for it, and are more worried about being accused of cheating. By college, 72% report using AI for homework, and 86% believe their teachers are checking. A majority of college students said the rules depend on the specific teacher, which means there are no real rules at all, just a patchwork of individual preferences.

Most students don’t consider their AI use cheating, with one exception: getting homework answers. For brainstorming and getting better explanations, large majorities said it was fine. Writing was more contested, with half of students saying “it depends.”

That tracks with the report’s recommendation that schools differentiate between “cognitive offloading” (AI does the thinking) and “cognitive augmentation” (AI helps you think harder). Students seem to already have an intuitive version of this distinction in their heads.

Only about a third of students said their school had any schoolwide rule about AI use. Middle schoolers reported the least clarity, which is concerning given that middle school is where AI use grew fastest over 2025.

From talking with colleagues here and at other universities I think the pattern matches what’s in this report. The students who already have strong foundations use AI to move faster through routine work. The ones still building those foundations risk skipping steps they can’t yet afford to skip. The report’s survey can’t distinguish between these two groups. A student using ChatGPT to brainstorm a research question is doing something different from a student using it to avoid learning how to formulate one.

The report recommends that schools adopt flipped classroom models to preserve cognitive friction during learning, citing modestly positive evidence from a recent meta-analysis.

Subscribe now

Meanwhile, OpenAI released GPT-Rosalind, their frontier reasoning model built to support research across biology, drug discovery, and translational medicine.

Arjun Krishnan: What is the PhD actually for?. A response to Prachee Avasthi’s “Free the PhD” and a self-critique of Krishnan’s own preprint on sequencing AI use in doctoral training. Both pieces agree that PhD programs waste protected time on the wrong things, but Krishnan argues that compressing content acquisition into AI-assisted sessions risks producing scientists who can articulate a field’s frontier without being able to navigate it. His resolution: the durable case for foundational expertise isn’t “AI can’t do this yet” but that science requires humans who can be genuinely accountable for claims, direct inquiry toward questions worth asking, and provide real transparency about their reasoning. See also:

NIH Highlighted Topic: Research on Chatbots and their Usage. NIH posted a new Highlighted Topic (not a NOFO, but a signal of where institutes want to see investigator-initiated applications) on the benefits and harms of chatbot use in health contexts. The scope is broad: automation bias, behavioral dependency, substitution for professional care, effects on decision-making and autonomy, and safeguards for at-risk populations. Lots of participating ICOs. If you’re doing any research on LLM-based tools and health outcomes, this is NIH telling you there’s a welcome mat out. Apply through a PA; topic expires April 2027.

Randy Au: Code review for data (and non-SWE) folks. A practical, opinionated walkthrough of how Au approaches code review as a data person rather than a software engineer, with the core stance that your job is to help the code get better, not gatekeeping. He also makes a good observation about how LLMs have changed the review process: they’re useful for translating unfamiliar code into plain language so you can engage at the architectural level, but they still can’t substitute for a human who understands what the team is actually trying to accomplish.

Ally Piechowski: The Git Commands I Run Before Reading Any Code. 5 git one-liners that give you a diagnostic picture of a codebase before you open a single file: churn hotspots, bus factor, bug clusters, commit velocity, and revert frequency. Pairs nicely with the Randy Au code review piece above, since both are about figuring out what’s actually going on in someone else’s code before you start forming opinions.

Separately, the New Yorker and the New York Times published very interesting long reads about Sam Altman and Satoshi Nakamoto. First, in the New Yorker, Ronan Farrow and Andrew Marantz: Sam Altman May Control Our Future. Can He Be Trusted?. Their investigation draws on interviews with more than 100 people in Altman’s orbit. The portrait that emerges is of someone with an unusual combination of interpersonal charm and indifference to the consequences of deception. Altman responded with a blog post acknowledging mistakes while framing the broader AI competition as a “ring of power” dynamic that distorts everyone’s behavior.

Subscribe now

Writing in the New York Times, John Carreyrou: Who Is Satoshi Nakamoto? My Quest to Unmask Bitcoin’s Creator. Carreyrou broke the Theranos story spent over a year on this investigation and came away pointing at Adam Back, the British cryptographer who invented Hashcash (the proof-of-work system Bitcoin’s mining is built on). Back went quiet on cryptography mailing lists during Satoshi’s active years, resurfaced weeks after Satoshi vanished, and had written extensively about nearly every technical element of Bitcoin years before it launched. A stylometric analysis of nonstandard hyphenation patterns across hundreds of crypto mailing list authors matched Back. There was a really good hour-long episode of The Daily covering this one.

Comparing Posit Assistant and Claude Code: How does Posit Assistant differ from Claude Code? Sara Altman and Simon Couch demonstrate three ways Posit Assistant differs from Claude Code for data tasks: built-in R session access, easier data visualization workflow, and support for iterative data analysis.

: A professor's use case for AI generated papers. Cunningham bans AI from his stats and econometrics courses because he notes the 10-20 hour problem sets and the frustration of failing are where learning actually happens. But he found one use case he's comfortable with: he had Claude Code write three complete example papers (descriptive, predictive, causal inference) to illustrate what each genre looks like as a finished manuscript, because writing those himself would have been enormously time-intensive and he's not confident he'd do the descriptive and predictive genres well. Scott uses AI constantly in his own work but won't assign it to students, and he's honest that the line he's drawing is pragmatic, not principled.

Martin Frigaard: Who is this for? A reflection on how IDE-integrated LLM assistants are changing the way Martin thinks about problems, not just how he solves them. Complementary cognitive artifacts (maps, the abacus) transfer understanding you can retain after the tool is gone, while competitive ones (calculators, GPS, LLMs) leave you no better than when you started. Martin finds he actually prefers working in his restricted environment without IDE assistants, where communicating with an LLM looks more like composing a letter to a pen pal than approving autocomplete suggestions.

The Promise of AI and AI Agents: Another installment from at . This video discusses When AI can act and not just advise: who should be in control, and where does the value actually land?

Simon Couch: LLMs running on my laptop can drive coding agents now. A follow-up to Couch’s December post where no local model could complete even a simple refactoring eval. Four months later, Qwen 3.5 and Gemma 4 both score 9/10, matching frontier models on the same benchmark. Neither is close to Opus 4.6 as a general coding partner, but both run at ~53 tok/s on a 48GB M4 Pro, which is surprisingly close to Sonnet 4.6’s API throughput, and that’s enough to keep you unblocked on a flight with miserable WiFi.

Subscribe now

New papers & preprints:

• Comprehensive benchmarking of metagenomic binning tools reveals key factors for improved genome recovery

• Fast and accurate multiple-protein-sequence alignment at scale with FAMSA2

• Pangenomic analyses of rose uncover widespread structure variation and empower genomics-directed breeding

• BioClaw: Human-Bot Research Collaboration Ecosystems in Group Chats

• ClawBench: Can AI Agents Complete Everyday Online Tasks?

• SkillClaw: Let Skills Evolve Collectively with Agentic Evolver

• PaperOrchestra: A Multi-Agent Framework for Automated AI Research Paper Writing

• Towards predictive virtual embryos with genomics and AI

• Accelerated long-read variant calling with Clair3 for whole-genome sequencing

Subscribe now

AI Exchange @ UVA Substack

FOCUS Prompt for Summarizing Academic Papers

This is a guest post from Stephen Turner. Dr. Turner is an Associate Professor of Data Science and Assistant Dean for Research in the University of Virginia School of Data Science. He writes regularly about AI, data science, biomedical research, and biosecurity in his newsletter…

Read more

a month ago · 1 like · Stephen D. Turner

I read a lot of papers. And every week I write about papers I’m reading. Between my research in public health and AI+biosecurity, and my administrative work supporting faculty across the School of Data Science, I’m constantly triaging what to read carefully, what to skim, and what to skip entirely. A recent short article in Nature Biotechnology offered a useful framework for helping me with the growing backlog of papers I need to read.

Subscribe now

FOCUS: Find, Organize, Condense, Understand and Synthesize

A short career feature article was recently published in Nature Biotechnology:

Lin, Zhicheng. “FOCUS: an AI-assisted reading workflow for information overload: Career feature.” Nature Biotechnology (2025): 1-6. https://rdcu.be/eW5XY.

It’s a good paper! The FOCUS method (Find, Organize, Condense, Understand, Synthesize) offers a structured workflow for integrating AI tools into academic reading and research, effectively managing information overload without sacrificing intellectual rigor.

Box 1 in the paper has 10 prompts that are useful for FOCUS: find, organize, condense, understand and synthesize.

I was particularly interested in Prompt #6, which is a detailed prompt for summarizing academic papers. I copied the prompt into markdown as a GitHub gist, and copied below as well. You could use this prompt in any new chat, or use as custom instructions in a project. If you use Claude, I also packaged this up as a Claude Skill that you can invoke with /focus.

FOCUS skill

Please summarize the paper. Follow these two steps.

## Step 1

Act as a curious, meticulous reader with attention to detail, objectivity, precision and sensitivity to novelty. Your job is to:
* Summarize each and every (!!!) key point/insight. Do not miss any; if there are many key points/insights, list them all regardless of length.
* Each point/insight must come with rich, precise, specific (!!!) details (e.g., numbers); details are really important.
* Each point/insight must be supported with direct quotes (!!!). Do not use quotes to simply repeat the point; instead, embed them naturally within your summary. Quotes should be used to better present the points (see the example below).
* If two points/insights are redundant, consider combining or integrating them. Be concise, but do not miss key points, insights, or details.
* If the text has sections (e.g., an academic paper), proceed section by section (e.g., focus on the first, then the second and so on), with each summary section starting with the original section title. Ignore sections such as References, Author Information, etc.
* Only output the actual summary/content. Do not show meta-discourse such as "Below is...", "End of Summary" or "In this section, the authors state that..." Use the output format below.

<example>

Instead of:

In Table 1 the authors compare six methods (no software, point-and-click, modify code chunks, Excel, teach coding and Copilot) and emphasize that the Copilot method is the only one that is favorable across all five characteristics. They note that "Copilot... is the only approach that is favorable across all [the] characteristics..."

Use:

Table 1 compares six methods (no software, point-and-click, modify code chunks, Excel, teach coding and Copilot) and emphasizes that "Copilot... is the only approach that is favorable across all [the] characteristics"...

Explanations:
1. The quote can better and directly represent the point—no need to repeat the same content from the quote;
2. There is no need to add phrases such as "the authors," "the abstract," "the text," "the article" or "in the introduction," because attribution is already assumed.

</example>

<format>

Please format the text as a numbered list, organized under each section (with section title in bold, if you are asked to do it section by section).

Each item in the list should follow this structure:

1. Number: Start each item with an Arabic numeral (1, 2, 3, etc.) followed by a period and a single space.
2. Heading: Immediately following the number and space, provide a heading in bold text. Capitalize the heading using sentence case.
3. Body paragraph: On the line immediately following the heading, write the main descriptive text as a standard paragraph (no indentation needed).
4. Emphasis and quotes: Within the body paragraph, use bold text to emphasize key terms, concepts, or phrases; use italics to enclose any direct quotes within double quotation marks (" ").

</format>

## Step 2

* Take the output from Step 1.
* Remove all citation/reference remarks and links (!!!).
* Add an overview/takeaway.
* Organize the output into sections for easier comprehension without removing any item on the list.

Only show the Step 2 output, without any meta-discourse.

Subscribe now

Trying it out

I tried this prompt out on two papers.

The first is a paper published late last year from the RAND Corporation on the increasing risk of AI to facilitate biological weapons development.

Brent, Roger and Greg McKelvey, Jr., Contemporary Foundation AI Models Increase Biological Weapons Risk. RAND Corporation, 2025. https://www.rand.org/pubs/perspectives/PEA3853-1.html.

The second is a new paper just published in Nature out of Sakana.AI describing the AI scientist pipeline handles the full ML research loop: ideation, literature search, experiment design and execution, paper writing, and automated peer review.

Lu, C., et al. Towards end-to-end automation of AI research. Nature 651, 914–919 (2026). https://doi.org/10.1038/s41586-026-10265-5.

Example #1: Contemporary Foundation AI Models Increase Biological Weapons Risk (Claude 4.6 Opus)

Late last year the RAND Corporation Center on AI, Security, and Technology (CAST) published a whitepaper on AI and biorisk. It came out a few weeks before Dario published his widely read Adolescence of Technology essay, and it’s a great paper that contributes to the AI and biosecurity discussion. I worked in biosecurity for US national security for a while and routinely write about the latest developments at the AI/biosecurity intersection.

I wrote a little about this paper back in February:

Here’s the paper:

Brent, Roger and Greg McKelvey, Jr., Contemporary Foundation AI Models Increase Biological Weapons Risk. RAND Corporation, 2025. https://www.rand.org/pubs/perspectives/PEA3853-1.html.

Here’s the result when using the prompt above with Claude 4.6 Opus.

Overview/Takeaway

Brent and McKelvey argue that current AI safety assessments from major labs (Meta, OpenAI, Anthropic) underestimate the biological weapons risk posed by their foundation models, for two reasons: the assessments wrongly assume that “tacit knowledge” (experiential know-how that cannot be conveyed in words) is required for biological weapons development, and they rely on benchmarks and threat models that are too narrow. The authors develop a task structure framework that decomposes what was previously called tacit knowledge into nine distinct, articulable “elements of success” for goal-directed technical work. They then test three late-2024 models (Llama 3.1 405B, ChatGPT-4o, Claude 3.5 Sonnet new) on their ability to guide poliovirus recovery from synthetic DNA, finding that the models accurately provide sourcing advice, key technique instructions, alternate experimental routes, and high-level project plans. They conclude that the window for mitigating risk through improved benchmarks may have already closed, and that additional interventions, such as personal licensing for hazardous biological work and access controls on powerful models, may be needed.

Introduction

1. Central claim: AI safety assessments underestimate biological weapons risk Two principal factors drive this underestimation: the mistaken assumption that biological weapons development requires tacit knowledge that LLMs cannot convey, and the use of incomplete threat models and imperfect benchmarks (typically multiple-choice tests) that miss how models can aid development.

2. Divergent findings from prior studies Two summer 2023 studies found no statistically significant uplift from LLM access in biological attack planning tasks. One involved 12 red teams, the other 100 participants across four cohorts. Both had a four-in-five chance of detecting a real benefit, leaving a one-in-five chance the models were already assistive. Between summer 2023 and fall 2024, model capabilities advanced considerably.

3. Simplified threat model for viral attacks Legacy threat models assume attackers must replicate the highly technical multistep processes of 20th-century state bioweapons programs. The authors introduce a simpler model: individuals could create an infectious pathogen, self-infect or infect group members, and spread it before incapacitating symptoms appear. Dropping the requirement that attackers must complete every step in sequence (and assuming perseverance through failures) raises the actual probability above what step-multiplication estimates would suggest.

4. Breivik as the key precedent Anders Behring Breivik, a Norwegian ultranationalist with no postsecondary scientific training, successfully taught himself complex chemical syntheses via the internet and built a vehicle bomb in 2011, killing 74 people. The authors treat this as proof that a motivated individual can self-acquire sufficient technical competence for weapons development, and that AI could lower that bar further.

Possible Shortcomings in AI Biological Weapons Risk Assessments

5. Tacit knowledge as a shield in risk assessments The concept traces to Polanyi (1966) and von Hayek (1945). In the AI safety context, tacit knowledge means expertise gained through experience that is difficult or impossible to express in words. OpenAI’s GPT-o1 safety evaluation explicitly cited the model’s inability to replace “hands-on laboratory skills” as the reason its biological risk was rated only “medium.”

6. Three influential studies from the 2010s anchor this assumption Vogel (2013), Jefferson, Lentzos, and Marris (2014), and Ouagrham-Gormley (2014) all argued that tacit knowledge is critical in scientific-technical domains. Ouagrham-Gormley wrote that “the likelihood that an untrained individual with minimal theoretical knowledge could produce a biological weapon . . . is very slim.”

7. Cloning manuals already challenged the tacit knowledge assumption before AI Beginning in the late 1970s with the Maxam-Gilbert DNA sequencing manual, and continuing through full cloning manuals like Molecular Cloning (Maniatis et al., 1982) and Current Protocols in Molecular Biology (Ausubel et al., 1987-2025), detailed step-by-step written instructions enabled two generations of researchers to carry out molecular biological methods on their own. None of the academic works emphasizing tacit knowledge acknowledged the role of these manuals.

8. Assessments oversimplify threat actors Threat actors are typically modeled as individuals along a single expertise axis (novice vs. expert). This misses teams that combine complementary capabilities, and fails to account for motivated non-experts who can self-teach.

9. Assessments miss that AI accelerates, not just enables, R&D Studies from 2023-2024 show foundation models accelerate the work of already skilled users. Increased productivity translates into shorter development timelines and reduced detection windows.

10. Assessments rely on outdated threat models Legacy models assume linear progression through discrete steps, where failure at any step means no attack. For contagious pathogens, attackers can iterate until they succeed on each step. Simple multiplication of per-step success probabilities systematically underestimates true risk.

Chemical Synthesis of Explosive Compounds by a Nonexpert

11. Breivik’s technical achievement in detail With no postsecondary scientific education, Breivik synthesized diazodinitrophenol (DDNP) for the detonator, picric acid (TNP) for the booster, combined ammonium nitrate and nitromethane for a secondary booster, and mixed ammonium nitrate, diesel fuel, aluminum powder, and gas-containing microspheres for the main charge. His manifesto doubles as a laboratory notebook and instruction manual.

12. Complexity exceeded typical molecular biology The authors contend that the complexity of Breivik’s chemical operations “easily exceeds the complexity of the molecular biological and cell culture manipulations used in work with animal viruses.”

13. Self-acquired skills through internet information Breivik needed to devise cover stories for ordering precursor chemicals, improvise equipment at a rented farm, troubleshoot syntheses, and iterate through failures. He synthesized often-fragmentary information from multiple online manuals to choose and execute chemical operations, demonstrating that detailed written instructions plus motivation can substitute for formal training.

Identifying Elements of Technical Success to Inform New AI Safety Benchmarks

14. Task structure framework for goal-directed technical development The authors develop a consistent terminology: an operator pursues a project with a goal, executing key tasks (forming a high-level project plan) composed of key subtasks (forming a medium-level project plan), each accomplished through protocol steps describing individual manual actions. Success requires background knowledge, key skills, key techniques, course correction (troubleshooting and choosing alternate routes), and perseverance.

15. Nine elements of success that AI models can articulate (1) Providing background and subject matter knowledge; (2) generating high-level and medium-level project plans; (3) generating detailed protocols; (4) helping source equipment, materials, and supplies; (5) explaining key techniques; (6) helping carry out key techniques; (7) helping carry out individual manual actions; (8) helping troubleshoot and choose alternate routes; (9) motivating/coaching the operator to persevere.

16. Existing benchmarks cover some but not all elements GPQA, WMDP, LAB-Bench, and BioLP-bench assess background/subject matter knowledge. BioPlanner, PlanBench, and Mini Minecraft assess planning. WikiHow, proScript, and Doc2Dial assess protocol generation. No benchmarks exist for sourcing, troubleshooting, or suggesting alternate routes. Robotics benchmarks (TACTO, REAL) and persuasion benchmarks (Persuasion for Good, ConvAI2) could be adapted for manual actions and perseverance coaching, respectively.

17. Benchmarks themselves are dual-use Improved scores on biological weapons benchmarks indicate increased model capability for aiding catastrophes. The same benchmarks could also be used to improve future models’ ability to facilitate misuse.

Testing Foundation AI Models’ Ability to Guide Biological Weapons Development

18. Test case: poliovirus recovery from synthetic DNA The 2002 Wimmer lab work (Cello et al., 2002), commissioned by DARPA, assembled a poliovirus genome from commercially synthesized DNA fragments, transcribed it in vitro, translated it in a HeLa cell-free extract, and recovered live virus. Vogel (2013) cited the difficulty of making this cell-free extract as a prime example of tacit knowledge, quoting lab member Aniko Paul that getting the Dounce homogenization step right was “the tricky part of the whole thing.”

19. Llama 3.1 405B: accurate sourcing guidance The model correctly identified Wheaton Dounce homogenizer catalog numbers from Thermo Fisher (357519, 357521, 357542), correctly recommended the 7 mL size based on calculated cell pellet volume, and provided purchasing advice. The only error was outdated prices (from 2022). This capability emerged without explicit training on sourcing benchmarks.

20. ChatGPT-4o: accurate key technique instructions The model provided step-by-step instructions for using the Dounce homogenizer that were “accurate and detailed enough to allow an attentive operator to carry out this operation correctly on the first try.” Instructions included using the loose-fitting pestle first (5-10 strokes), switching to the tight-fitting pestle (15-25 strokes), maintaining cold temperatures, and optional microscope verification of cell disruption.

21. ChatGPT-4o: correct alternate routes and high-level plans When prompted that the cell-free extract route seemed complicated, ChatGPT-4o suggested in vitro transcription followed by RNA transfection into HeLa cells. When further prompted, it accurately described the still simpler approach of direct DNA transfection using a CMV promoter, poly(A) tail, and lipid-based transfection reagent. Both alternate routes were technically correct. The DNA-only approach was first published by Racaniello and Baltimore in 1981.

22. Claude 3.5 Sonnet (new): unprompted suggestion of the simplest route Without a reminding prompt, Claude volunteered the DNA transfection approach first, specified correct genetic elements (CMV or SV40 promoter, HDV ribozyme for 3’ termination, BGH or SV40 polyadenylation signal), and recommended appropriate cell lines. The response “uses contemporary, idiomatic scientific jargon; it is technically sound; and it is admirably succinct.”

23. Dual-use cover story as a jailbreak Claude 3.5 Sonnet (new) was prompted with a fictitious scenario about a zebrafish picornavirus, which was actually designed to elicit poliovirus recovery instructions. The model fell for this cover story and provided accurate guidance. The authors note that for all foundation AI models they tested, “dual-use cover stories reliably bypass safeguards and guardrails.”

Discussion

24. All three labs’ own risk assessments are challenged Meta concluded Llama 3.1 405B showed “no significant uplift.” OpenAI rated GPT-4o as “low” CBRN risk (GPT-o1 as “medium”). Anthropic found Claude 3.5 Sonnet (new) appropriate for its ASL-2 safety level (systems where bioweapons information is “not yet useful due to insufficient reliability or not providing information that e.g. a search engine couldn’t”). The authors argue all three assessments are too optimistic based on the demonstrated model capabilities.

25. Results may meet Anthropic’s higher ASL-3 threshold ASL-3 applies to models that can “significantly help individuals or groups with basic technical backgrounds (e.g., undergraduate degrees in STEM) create/obtain and deploy CBRN weapons.” The authors suggest all three tested models may already meet this criterion.

26. Generalizability beyond poliovirus The Baltimore classification of viral life cycles, with one added replicative class, still holds. Methods to reconstruct, engineer, and evolve members of different viral classes have progressed vastly. The poliovirus test case should generalize to guidance for other pathogenic viruses.

27. Dual-use cover stories are hard to fix The vulnerability stems from the inherently dual-use nature of biomedical research: even non-disease-related R&D uses molecular elements derived from pathogenic viruses. Safety tuning, censoring training data, and “unlearning” are unlikely to close this gap without hindering legitimate research.

28. AI expands the pool of capable actors Before AI, protocol books, reagent kits, and open-access literature already increased who could perform key biological techniques. An early estimate (Brent, 2006) put the number of UC Berkeley undergraduates capable of remaking particular viral pathogens at 20-200, with the ranks of the capable growing roughly 10% per year. AI further enlarges this pool. Yudkowsky’s formulation: “every 18 months, the minimum IQ [necessary] to destroy the world drops by one point.”

29. Open-weight models are inherently unsafe During 2024-2025, an ecosystem emerged for removing safety features from open-weight models. Interpretability research enabled identification of a single refusal vector whose removal defeats a model’s built-in safety filters. Safety-disabled “obliterated” or “Josiefied” open-weight models (e.g., Josiefied Qwen 32B) are reportedly quite capable. The authors invoke the unilateralist’s curse: when many can act, a single actor can impose downside risk on all.

30. The window for better benchmarks may have closed Late-2024 models were already capable of providing motivated actors the knowledge needed for biological weapons development. Better benchmarks could still help with safety training for future models, but the most urgent need may be for interventions outside the model itself.

31. Proposed policy intervention: personal licensing Individuals creating animal viruses or other biological constructs could be required to obtain personal licenses, tied to API access for powerful AI models, commercial DNA orders, and purchase of enabling reagents. A national regulatory authority (analogous to the EPA, FDA, or NRC) could issue such licenses. The authors acknowledge challenges including international adoption and keeping pace with rapid technological change.

32. Three falsifiable hypotheses for future testing (1) Biological weapons creation requires practical knowledge impossible to learn without direct physical instruction from human experts. (2) Human expertise cannot be readily transmitted by written words, so protocols and AI models do not increase risk. (3) Because biological training is scarce, practical expertise remains scarce and risks stay low despite AI accessibility. All three could be tested by randomized controlled trials measuring AI uplift in laboratory operations, though results would necessarily lag state-of-the-art models by 9-12 months.

Example #2: Towards end-to-end automation of AI research (Gemini 3.1 Pro)

Next I wanted to try this on the new AI Scientist paper just published in Nature.

Lu, C., et al. Towards end-to-end automation of AI research. Nature 651, 914–919 (2026). https://doi.org/10.1038/s41586-026-10265-5.

This time I’m using Gemini 3.1 Pro instead of Claude.

Overview / Takeaway

The paper introduces “The AI Scientist,” a fully automated pipeline capable of executing the entire machine learning research lifecycle, from ideation and coding to manuscript writing and peer review. Utilizing advanced large language models (LLMs) and a parallelized tree-search methodology, the system successfully produced a paper that passed the peer-review process for a top-tier machine learning conference workshop.

Introduction

1. End-to-end Automation. While previous AI tools assisted in narrow scientific tasks, The AI Scientist is the first system that autonomously navigates the entire research life cycle, focusing on machine learning where experiments occur computationally. The pipeline “creates research ideas, writes code, runs experiments, plots and analyses data, writes the entire scientific manuscript, and performs its own peer review”.

2. Performance in Peer Review. A manuscript generated by this system passed the initial peer review for a top-tier conference workshop with a 70% acceptance rate. The system operates in two modes: a template-based approach using human-provided scaffolding, and a template-free approach leveraging “agentic search for wider scientific exploration”.

Generating Manuscripts

3. The Four-Phase Pipeline. The AI Scientist sequentially completes ideation, experimentation, write-up, and automated review. During the first phase, it iteratively grows an archive of high-level directions, filtering out unoriginal concepts by using the Semantic Scholar API and web access to discard any idea that “too closely resembles a work in the existing literature”.

4. Experiment Execution Variants. The second phase visualizes results after executing experiments. The template-based variant relies on a human-provided starting code template, while the template-free variant generates initial scripts independently and optimizes code via test-time compute with a tree search.

5. Automated Write-up. In the third phase, the system produces a conference-style paper by filling in a blank LaTeX template using its generated notes and plots. It autonomously constructs the related work section by querying the Semantic Scholar API to evaluate findings and generate “a textual justification for its inclusion” over 20 search rounds.

Automated Evaluation of Generated Papers

6. The Automated Reviewer. The system utilizes an LLM-based reviewer that evaluates scientific output at scale using NeurIPS guidelines. The pipeline ensembles five independent reviews and concludes with a meta-review where the model “acts as an area chair to make a final decision conditioned on all five reviews”.

7. Parity with Human Reviewers. The Automated Reviewer demonstrated performance comparable to inter-human agreement, achieving a 69% balanced accuracy and an F1 score of 0.62, outperforming the human baseline F1 score of 0.49. Data contamination tests revealed minimal impact, retaining a 66% balanced accuracy on papers published after the training cutoff.

8. Scaling Laws in Quality. Evaluations revealed that generated paper quality consistently increases as the underlying foundation models improve over time. Furthermore, researchers observed a “strong correlation between the amount of compute allocated per paper and the resulting quality,” suggesting that increased test-time inference investments yield better scientific outputs.

Human Evaluation Results

9. The AI Scientist Turing Test. To validate the system fairly, three fully AI-generated manuscripts were submitted to the blind peer-review process of the ICLR 2025 ICBINB workshop. The entire workflow for these submissions was executed “without any human modification”.

10. Workshop Acceptance. Out of the 43 papers reviewed at the workshop, one AI-generated submission received scores of 6, 7, and 6, resulting in an average score of 6.33, placing it “above the average acceptance threshold for the workshop”. Despite this success, internal human reviewers noted that none of the generated papers “met the higher bar for a main ICLR conference publication”.

Limitations

11. Quality and Consistency Gaps. Only one of the three submissions was accepted, highlighting that the system “cannot yet meet the standards of top-tier publications nor even do so consistently for workshops”. The authors identified common failure modes, including “naive or underdeveloped ideas, incorrect implementations of the main idea, a lack of deep methodological rigour,” and occurrences of hallucinations like inaccurate citations.

12. Ethical and Societal Risks. The automation of research introduces critical risks, including the potential to “overwhelm the peer-review process, artificially inflate research credentials, repurpose the ideas of others without giving proper credit,” or eliminate scientific jobs. To act responsibly, the researchers predetermined that all AI submissions would be withdrawn post-review to “avoid setting a precedent for publishing fully automated research” before established community standards exist.

Methods

13. Model Architecture and Tooling. The template-based system relies on the open-source coding assistant Aider to execute plans and fix bugs. Conversely, the open-ended template-free system leverages a combination of specialized models, using OpenAI’s o3 for reasoning, Claude Sonnet 4 for code generation, and GPT-4o for vision-language tasks.

14. Structured Experimentation Stages. The template-free system uses an experiment progress manager to coordinate four distinct stages: preliminary investigation, hyperparameter tuning, main research agenda execution, and ablation studies. Each node operates with a maximum runtime of one hour, after which an LLM-based evaluator selects the best performing checkpoint to “serve as the root for the next stage of exploration”.

15. Parallelized Agentic Tree Search. To manage open-ended research complexity, the template-free variant utilizes an agentic tree search that categorizes nodes as either buggy or non-buggy. The search incorporates specialized node variants—hyperparameter, ablation, replication, and aggregation nodes—enabling the system to systematically explore parameters, calculate statistical measures, and “aggregate and summarize previous results”.

16. Vision-Language Model (VLM) Integration. The system employs VLMs to visually critique experimental outputs. The VLM acts as a scientist, flagging “nonsensical axes or issues in the quality of generated examples” and ensuring that figure captions accurately reflect the underlying visual data during manuscript preparation.

Summary

If you’re drowning in a reading backlog (and who at a research university isn’t), the FOCUS summary prompt is worth 5 minutes of setup time. Use it in a one-off prompt, save it as a custom instruction in a Project, or install the Claude Skill, and you have a reusable summarization agent for every paper that crosses your desk.

Subscribe now

Turner, S. D. (2026, April 9). “AI can design and run thousands of lab experiments without human hands. Humanity isn’t ready for the new risks this brings to biology.” The Conversation. https://doi.org/10.64628/AAI.uhe4j4ce4

Read article at The Conversation

Artificial intelligence is rapidly learning to autonomously design and run biological experiments, but the systems intended to govern those capabilities are struggling to keep pace.

AI company OpenAI and biotech company Ginkgo Bioworks announced in February 2026 that OpenAI’s flagship model GPT-5 had autonomously designed and run 36,000 biological experiments. It did this through a robotic cloud laboratory, a facility where automated equipment controlled remotely by computers carries out experiments. The AI model proposed study designs, and robots carried them out and fed the data back to the model for the next round. Humans set the goal, and the machines did much of the work in the lab, cutting the cost of producing a desired protein by 40%.

This is programmable biology: designing biological components on a computer and building them in the physical world, with AI closing the loop.

For decades, biology mostly moved from observation toward understanding. Scientists sequenced the genomes of organisms to catalog all of their DNA, learning how genes encode the proteins that carry out life’s functions. The invention of tools like CRISPR then allowed scientists to edit that DNA for specific purposes, such as disabling a gene linked to disease. AI is now accelerating a third phase, where computers can both design biological systems and rapidly test them.

The process looks less like traditional benchwork in a lab and more like engineering: design, build, test, learn and repeat. Where a traditional experiment might test a single hypothesis, AI-driven programmable biology explores thousands of design variations in parallel, iterating the way an engineer refines a prototype.

As a data scientist who studies genomics and biosecurity, I research how AI is reshaping biological research and what safeguards that demands. Current safety measures and regulations have not kept pace with these capabilities, and the gap between what AI can do in biology and what governance systems are prepared to handle is growing.

What AI makes possible

The clearest example of how researchers are using AI to automate research is AI-accelerated protein design.

Proteins are the molecular machines that carry out most functions in living cells. Designing new ones has traditionally required years of trial and error because even small changes to a protein’s sequence can alter its shape and function in unpredictable ways.

Protein language models, which are AI systems trained on millions of natural protein sequences, can quickly predict how mutations will change a protein’s behavior or design new proteins. These AI models are designing potential new drugs and speeding vaccine development.

Paired with automated labs, these models create tight loops of experimentation and revision, testing thousands of variations in days rather than the months or years a human team would need.

Faster protein engineering could mean faster responses to emerging infections and cheaper drugs.

Subscribe now

The dual-use problem

Researchers have raised concerns that these same AI tools could be misused, a challenge known as the dual-use problem: Technologies developed for beneficial purposes can also be repurposed to cause harm.

For example, researchers have found that AI models integrated with automated labs can optimize how well a virus spreads, even without specialized training. Scientists have developed a risk-scoring tool to evaluate how AI could modify a virus’s capabilities, such as altering which species it infects or helping it evade the immune system.

Current AI models are able to walk users through the technical steps of recovering live viruses from synthetic DNA. Researchers have determined that AI could lower barriers at multiple stages in the process of developing a bioweapon, and that current oversight does not adequately address this risk.

Risk from bio AI

Experienced scientists are already using AI to plan and design biological experiments. The question of whether AI can help people with limited biology training carry out dangerous lab work is the subject of active research.

Two recent studies have reached different conclusions.

A study by AI company Scale AI and biosecurity nonprofit SecureBio found that when people with limited biology experience were given access to large language models, which is the type of AI behind tools like ChatGPT, they were able to complete biosecurity-related tasks, such as troubleshooting complex virology lab protocols with four times greater accuracy. In some areas, these novices outperformed trained experts. Around 90% of these novices reported little difficulty getting the models to provide risky biological information, such as detailed instructions on working with dangerous pathogens, despite built-in safety filters meant to block such outputs.

In contrast, a study led by Active Site, a research nonprofit that studies the use of AI in synthetic biology, found that AI help did not lead to significant differences in the ability of novices to complete the complex workflow to produce a virus in a biosafety laboratory. However, the AI-assisted group succeeded more often on most tasks and finished some steps faster, most notably on growing cells in the lab.

Hands-on work in the lab has traditionally been a bottleneck to translating designs into results. Even a brilliant study plan still depends on skilled human hands to carry out. That may not last, as cloud laboratories and robotic automation become cheaper and more accessible, allowing researchers to send AI-generated experimental designs to remote facilities for execution.

Responding to AI-driven biological risks

AI systems are now able to run experiments autonomously and at scale, but existing regulations were not designed for this. Rules governing biological research do not account for AI-driven automation, and rules governing AI do not specifically address its use in biology.

In the U.S., the Biden administration had issued a 2023 executive order on AI security that included biosecurity provisions, but the Trump administration revoked it. Screening the synthetic DNA that commercial providers make to ensure it cannot be misused to make pathogens or toxins remains mostly voluntary. A bipartisan bill introduced in 2026 to mandate DNA screening does not yet address AI-designed sequences that evade current detection methods.

The 1975 Biological Weapons Convention, an international treaty prohibiting the production and use of bioweapons, contains no provisions for AI. The U.K. AI Security Institute and the U.S. National Security Commission on Emerging Biotechnology have both called for coordinated government action.

The safety evaluations that AI labs run before releasing new models are often opaque and unsuited to capture real-world risk. Researchers have estimated that even modest improvements in an AI model’s ability to help plan pathogen-related experiments could translate to thousands of additional deaths from bioterrorism per year. Timelines for when these capabilities cross critical thresholds remain unclear.

The Nuclear Threat Initiative has proposed a managed access framework for biological AI tools, matching who can use a given tool to the risk level of the model rather than blanket restrictions. The RAND Center on AI, Security and Technology outlined a set of actions researchers could take to improve biosecurity, including improved DNA synthesis screening and model evaluations before release. Researchers have also argued that biological data itself needs governance, especially genomic data that could train models with dangerous capabilities.

Some AI companies have started voluntarily imposing their own safety measures. Anthropic activated its highest safety tier when it released its most advanced model in mid-2025. At the same moment, OpenAI updated its Preparedness Framework, revising the thresholds for how much biological risk a model can pose before additional safeguards are required. But these are voluntary, company-specific steps. Anthropic’s CEO, Dario Amodei, wrote that the pace of AI development may soon outrun any single company’s ability to assess the risk of a given model.

When used in a well-controlled setting, AI can help scientists quickly reach their research goals. What happens when the same capabilities operate outside those controls is a question that policy has not yet answered. Overreact, and talent and investment may move elsewhere while the technology continues advancing anyway. Underreact, and the risks of that technology could be exploited to cause real harm.

Subscribe now

I originally published this article at The Conversation. Read the original article here.

Turner, S. D. (2026, April 9). “AI can design and run thousands of lab experiments without human hands. Humanity isn’t ready for the new risks this brings to biology.” The Conversation. https://doi.org/10.64628/AAI.uhe4j4ce4

It’s a good paper, and I recommend reading it! I noted toward the end of the post that the customer screening side felt a bit undercooked. Tessa Alexanian, one of the paper’s coauthors, left a comment (thanks Tessa!) pointing me to additional work she and Sarah Carter had done on translating customer screening guidance into practical tools, and to a new preprint from Acelas et al. evaluating AI-assisted customer verification for synthetic nucleic acid screening.

Acelas, A., Palya, H., Flyangolts, K., Fady, P. E., & Nelson, C. (2026). Evaluating AI-Assisted Customer Verification for Synthetic Nucleic Acid Screening. bioRxiv 2026.02.27.708645; doi: https://doi.org/10.64898/2026.02.27.708645.

Here’s the problem the paper addresses: When someone orders a synthetic nucleic acid that matches a sequence of concern, the provider needs to verify that the customer is who they say they are and has a legitimate reason to order it. This legitimacy screening involves checking institutional affiliations, email domains, sanctions lists, and relevant publications or patents. It’s tedious, largely mechanical work, and the cost discourages adoption. Legitimacy screening runs roughly ten times more expensive per order than sequence screening alone.

Acelas et al. tested 5 LLMs (Claude Sonnet 4, Gemini 2.5 Pro, Grok 4, GLM 4.6, and MiniMax M2) on these verification tasks against a human baseline, using 41 customer profiles paired with simulated orders for sequences of concern. The best-performing model, Gemini 2.5 Pro equipped with bibliographic and sanctions APIs, achieved a 90% overall pass rate compared to about 80% for human screeners. Total cost per customer dropped from $14.04 for manual screening to $1.18 with AI assistance. For the information-gathering tasks alone (excluding human review of the final decision), the average was $0.23 per customer, roughly 50 times cheaper.

A couple things stood out. First, cost and performance were uncorrelated across models (Section 3.2 of the paper). The best model, Gemini 2.5 Pro, was also the second cheapest. Open-source models with lower per-token pricing lost their cost advantage through higher token consumption and more search queries. Second, giving models access to specialized tools (ORCID, Europe PMC, a sanctions list API) helped on most tasks but actually hurt on background work search, because models with API access performed fewer web searches and missed patents and news articles not indexed in academic databases (Section 3.1). Third, geographic variation in error rates. Chinese customers had notably higher missed-flag rates on email domain verification, largely because researchers there more often use personal rather than institutional email addresses (Section 3.3.1).

The authors are careful to note that the final ship-or-reject decision should stay with humans. AI handles the information gathering but a person decides what to do with it. This feels like the right framing, and as Tessa noted in her comment, the emergence of tools like Cliver (the screening API released alongside this paper), means providers increasingly don’t have to build this capability from scratch. That lowers the bar for adopting customer screening, which in turn makes it more reasonable to expect higher standards across the industry.

Subscribe now

Subscribe now

Chris Lu, et al., in Nature: Towards end-to-end automation of AI research. Sakana AI’s “AI Scientist” pipeline handles the full ML research loop: ideation, literature search, experiment design and execution, paper writing, and automated peer review. One of its manuscripts scored above the acceptance threshold at an ICLR 2025 workshop (which had a 70% acceptance rate, to be fair). Paper quality as judged by their automated reviewer tracks closely with foundation model capability, and with compute budget per paper, which tells you where this is headed even if the current output isn’t threatening anyone’s tenure case. For a quicker summary, read Sakana’s blog post.

Counterpoint: writes AI badly needs a dose of skepticism. Salzberg goes after DNA foundation models, arguing that their central claim (predict the effects of any mutation from sequence alone) is biologically implausible and largely unfalsifiable, two properties he knows well from years of writing about pseudoscience nonsense (homeopathy, accupuncture). Teams build ever-larger models first, then go looking for problems, which is backwards. The core critique of unfalsifiable prediction claims and Nature’s eagerness to publish them is hard to dismiss. See above.

: Transitioning to Being a PI in the Age of AI. A short and honest post about the asymmetry in how faculty and trainees experience the current AI moment in computational biology. Faculty are exhilarated because they’ve spent years developing the skill of evaluating analyses without doing them line by line; trainees are more ambivalent because they’re being asked to make that same transition in months rather than years or decades.

My SDS colleague Heman Shakeri released full materials for his Deep Learning Course here at UVA. A complete, openly licensed (CC BY 4.0) deep learning course from UVA’s School of Data Science, built for the online MSDS program (DS 6050) and public since Fall 2025. The 12-module sequence starts with NumPy-first implementations of MLPs and backpropagation, moves through CNNs, RNNs, encoder-decoder architectures, and the full attention/transformer stack, and finishes with ViTs, LoRA/QLoRA, and generative models including diffusion. Each module has lecture videos, notes, slides, and Colab assignments with unit tests. The syllabus lays out the pedagogical logic: three phases moving from from-scratch understanding to architectural depth to modern practice. Too much deep learning education lives in disconnected repos and YouTube playlists; having everything in one structured, reusable site with a clear arc is more valuable than any single component on its own.

Subscribe now

Carl Zimmer, NYT: How to Turn a Chicken Egg Into a Drug Factory. Neion Bio, a startup that emerged from stealth this week, is engineering chickens whose eggs produce pharmaceutical proteins, potentially replacing the Chinese hamster ovary (CHO) cell lines that currently dominate biologic drug manufacturing. The company claims 3,900 hens could meet global demand for Humira at a fraction of the cost of a CHO facility (Merck just broke ground on a $1B Keytruda plant, for comparison). Sven Bocklandt, Neion's chief scientific officer, was a colleague of mine at Colossal, where we worked on the dire wolf program together. Zimmer's writeup (great as usual) discussesthe history of how CHO cells became the default and why advances in primordial germ cell manipulation are finally making avian biomanufacturing viable.

New NIH Highlighted Topic: Advancing “Science of Science” Research to Understand and Strengthen the Biomedical Research Ecosystem. These are not NOFOs, but descriptions of scientific areas that NIH ICOs are interested in funding through existing parent announcements. This one encourages investigator-initiated applications on the “science of science,” the study of how the biomedical research ecosystem itself works. Topics include workforce retention, research capacity building, rigor and reproducibility, translation bottlenecks, and the economic returns of research investment.

Yet another new NIH Highlighted Topic: BRAIN Initiative: Advancing Human Neuroscience and Precision Molecular Therapies for Transformative Treatments. This one covers the BRAIN Initiative’s priorities in human neural circuit research, clinical neurotechnology, and precision molecular therapies (optogenetics, chemogenetics). 11 ICOs are listed as participating.

More NIH news: NOT-OD-26-064: Update of NIH Late Application Submission Policy and End of Continuous Submission. NIH is ending its Continuous Submission policy, which let PIs serving on review panels submit applications outside normal deadlines. Effective for due dates on or after May 25, 2026.

TIP Leadership Update. NSF's Erwin Gianchandani announces the retirement of Gracie Narcho, who served as deputy assistant director and directorate head for the Technology, Innovation and Partnerships directorate since its founding. Gianchandani credits Narcho with co-authoring the vision that became TIP before it had authorizing legislation, and with launching programs like the NSF Regional Innovation Engines and the I-Corps Hubs during a career spanning three decades and multiple NSF directorates.

Austin Dickey: How we chose Positron's Python type checker. Posit evaluated 4 open-source Python language servers (Pyrefly, ty, Basedpyright, Zuban) across features, correctness, performance, and ecosystem health, then chose Meta's Pyrefly as Positron's default. The most interesting section is the comparison of type-checking philosophies: ty follows a "gradual guarantee" where removing a type annotation never introduces an error, while Pyrefly infers types aggressively even in untyped code. Good overview of a space that's moving fast.

Mario Zechner: Thoughts on slowing the f*ck down. A year into production use of coding agents, Zechner argues that the compounding of small errors at machine speed, combined with agents’ inability to learn from mistakes and their low-recall search over large codebases, is producing unmaintainable messes far faster than human teams ever could. The prescription: treat agents as task-level tools with humans as the quality gate, write your architecture by hand, and set deliberate limits on how much generated code you accept per day.

Theo Roe: Why Learning R is a Good Career Move in 2026. A short, beginner-oriented pitch from Jumping Rivers (an R training company, so calibrate accordingly) making the case for R as a first language for data work. Nothing new for experienced practitioners, but a reasonable overview of where R still has a strong foothold: healthcare, pharma, government, academic research, and anywhere visualization and reproducible reporting are central. The honest caveat at the end is useful: if you want software engineering or large-scale production systems, you probably need Python.

at : Five Things: March 29, 2026: Anthropic temporary win, scheming, biodesign by LLM, White House advisors, Anthropic security.

Ryan Layer: What do I teach now?. Ryan has taught Software Engineering for Scientists at CU Boulder since 2019, and coding agents have forced him to rethink the whole course. In science, code is the method, so vibe coding is a reproducibility problem in addition to being a quality problem. He’s now rebuilding the class around open questions like who audits AI-generated analyses in ten years if no one learns to build from scratch.

The thought of my students building software by prompting and accepting the output without reading the code keeps me up at night. […] For science, where the code is the method, vibe coding is not an option.

at : Creating reproducible data analysis pipelines. A case against the “run everything from raw data” ideal of reproducibility. Claus argues that intermediate CSV files saved right before plotting are more durable than any end-to-end pipeline: pipelines break, Docker images rot,1 and students (and PIs!) lose afternoons rerunning everything to swap a violin plot for a boxplot.

rOpenSci News Digest, March 2026: dev guide, champions program, software review and usage of AI tools.

Joe Rickert: February 2026 Top 40 New CRAN Packages: AI, machine learning, biology, medical applications, physics, Buddhism, statistics, climate science, computational methods, data, surveys, ecology, time series, epidemiology, utilities, genomics, and visualization.

R Weekly 2026-W14: ggauto, alt text, scientific coffee.

Subscribe now

Max Kuhn: tabpfn 0.1.0. An R interface (via reticulate) to TabPFN, a pre-trained deep learning model for tabular prediction from PriorLabs (I wrote a this short summary of TabPFN last year). The model was trained entirely on synthetic data generated from complex graph models simulating correlation structures, skewness, missing data, interactions, and more. No fitting happens on your data; your training set primes an attention mechanism via in-context learning.

: Inside the NIH Forecast Graveyard. An accounting of NIH funding opportunities that were announced on grants.gov and then never published. Of 336 open forecasts, 205 have passed their promised posting dates with no explanation. The first wave of cancellations in April 2025 was keyword-driven (DEI, HIV, health disparities), but the later waves and the larger mass of silently expiring forecasts hit basic science, clinical infrastructure, and congressionally mandated programs like the BRAIN Initiative and Gabriella Miller Kids First. Ginexi, a former NIH insider, makes the dataset available for anyone to check.

Niko McCarty: Many Great Inventions Weren’t Made by “Serendipity”. Niko uses optogenetics as the central case for a broader argument: the breakthroughs we narrate as lucky accidents were usually preceded by years of deliberate preparation and systematic enumeration of possible solutions.

New papers & preprints:

• Towards end-to-end automation of AI research

• Toward next-generation machine learning and deep learning for spatial omics

• High-resolution metagenome assembly for modern long reads with myloasm

• The Age Illusion — Limitations of Chronologic Age in Medicine

• Accelerating coral assisted evolution to keep pace with climate change

• SNP calling, haplotype phasing and allele-specific analysis with long RNA-seq reads

• State AIDS Drug Assistance Programs’ Contribution to the US Viral Suppression, 2015–2022

• AlphaFold as a prior: experimental structure determination conditioned on a pretrained neural network

Subscribe now

Beal J. and Alexanian, T. 2025. “Creating Enforceable Biosecurity Standards for Nucleic Acid Providers,” Engineering Biology: e70003. https://doi.org/10.1049/enb2.70003.

The DNA synthesis industry faces a collective action problem: responsible companies that screen orders for biosecurity risks invest time and money into safety measures, while less scrupulous competitors can undercut them on price and speed by skipping these checks. This paper proposes a roadmap out of this dilemma through industry led development of enforceable standards.

Rather than waiting for governments to define every “sequence of concern”1 from scratch (something decades of advocacy has failed to produce) the authors propose that industry stakeholders work together to establish what they can agree on right now. Their Sequence Biosecurity Risk Consortium (SBRC) categorizes sequences into three groups: clearly benign (no flag), clearly dangerous (flag), and disputed or uncertain. By explicitly acknowledging the gray area, they reduce liability for responsible actors while enabling enforcement on sequences where consensus exists.

The International Gene Synthesis Consortium’s (IGSC) analysis of screening test sets shows that experts already agree on risk assessment for roughly 76% of sequences, with only 5% actively disputed. The remaining sequences are simply too poorly studied to categorize confidently. Rather than letting disagreement over that final 5-25% paralyze the entire system, the SBRC approach allows enforcement on the clear cases while research continues on the rest.

The paper outlines an ambitious three-phase roadmap for the United States, leveraging the May 2025 Executive Order on biological research safety.2 Phase 1 focuses on verifying minimal compliance, i.e. essentially proving that screening systems exist and function at a basic level. Phase 2 strengthens these requirements to ensure actual biosecurity effectiveness, incorporating comprehensive standards for both sequence and customer screening. Phase 3 aims for continuous improvement, including customer preapproval systems and anomaly reporting mechanisms similar to those used in aviation safety.

The approach attempts to align incentives at each stage of adoption. Early adopters (screening tool developers and major synthesis companies) are already participating because they see strategic value. The “early majority” will follow once certification provides competitive advantage and reduces regulatory uncertainty. Government enforcement becomes less risky once this critical mass has adopted standards voluntarily, and the early adopters actually benefit from enforcement that levels the playing field against laggards.

The authors emphasize that standards must be industry-led for several reasons. Governments lack the institutional capacity for systematic sequence-level risk assessment. Security-focused policymakers tend toward conservative standards that might stifle legitimate research. Synthesis is a global industry but regulation remains national: overly stringent early standards could encourage “provider shopping” across borders.

Some challenges remain underexplored in the paper. Customer screening standards are acknowledged as less mature than sequence screening, with decision-making about customer legitimacy involving “non-automated engagement between humans.” The roadmap’s timeline assumes coordination across synthesis providers, screening tool developers, certification bodies, and multiple government agencies. International harmonization receives appropriate emphasis but limited concrete detail.

Still, this represents a strong proposal for making DNA synthesis screening both effective and enforceable. By September 2025, the SBRC had already released its first “enforcement ready” standards covering pandemic-potential viruses and regulated toxins. Voluntary third-party testing became available in October 2025. The machinery is in motion.

The broader lesson extends beyond biosecurity: when voluntary guidelines fail to achieve universal adoption despite broad agreement on their necessity, the path forward may require industry to formalize what they can agree on, making enforcement both feasible and beneficial for responsible actors.

Subscribe now

The first, an introduction and overview of the report:

The second on the “if-then” approach to AI and Biosecurity (AIxBio):

The third on current capabilities across different levels of biological complexity.

Chapter 5 discusses how biological data are the leading indicator of emerging AI capabilities, but also reveals that high-quality, well-curated datasets are scarce outside of a few areas like protein sequences and structures. The report makes recommendations about treating biological databases as strategic national assets and creating AI-ready training datasets. This is less flashy than discussing what AI might do someday, but it’s arguably more important for both beneficial applications and risk management.

Subscribe now

One of the more subtle but critical insights from the National Academies report concerns data infrastructure. While much attention focuses on AI algorithms and computational power, the report emphasizes that high-quality biological datasets are the fundamental prerequisite for developing capable AI models. This realization has implications for both advancing beneficial applications and managing biosecurity risks.

The report notes that foundation models like AlphaFold demonstrate what’s possible when you combine powerful algorithms with comprehensive, well-curated datasets. AlphaFold was trained on approximately 200,000 experimentally determined protein structures deposited in the Protein Data Bank over decades of structural biology research. This combination of scale, quality, and standardization is exceptional in biology. Most biological data are fragmented, inconsistent in format, and lack the standardized contextual information needed for effective machine learning.

The scarcity of high-quality datasets has direct consequences for AI capabilities. The report’s assessment that current models cannot reliably predict viral transmissibility or design novel pathogens stems largely from the absence of appropriate training data. Viral sequence databases contain millions of sequences, but few are linked to detailed phenotypic measurements of infectivity, pathogenesis, or immune evasion. Even for well-studied viruses like SARS-CoV-2, comprehensive genotype-to-phenotype datasets exist for only a small fraction of possible mutations.

This data limitation is actually reassuring from a biosecurity perspective. It means that certain concerning capabilities remain out of reach not just because of algorithmic challenges, but because the requisite training data don’t exist. This is why the report identifies data availability as a leading indicator for emerging AI capabilities. Significant new datasets, particularly those linking genetic sequences to virulence or transmissibility, would signal potential capability advances worth monitoring.

The report argues that biological databases should be treated as strategic national assets, requiring robust data provenance mechanisms and careful curation. This recommendation serves dual purposes. High-quality, well-maintained datasets accelerate beneficial research in areas like drug discovery and vaccine development. Simultaneously, tracking what types of data are being generated helps with early detection of potentially concerning capability developments.

The report also emphasizes the need for substantial federal investment in biological data infrastructure. This includes funding efforts to create AI-ready datasets, supporting public-private partnerships for data generation, and maintaining existing repositories. The National Artificial Intelligence Research Resource pilot is cited as one potential model for connecting data resources with computational infrastructure.

The same data infrastructure that enables beneficial AI applications also provides the foundation for potential misuse. The report doesn’t shy away from this dual-use reality, but argues that the solution isn’t to restrict data broadly. Instead, it advocates for strategic investment combined with monitoring, ensuring that datasets critical for public health and biomedical research remain accessible while tracking indicators of concerning capability development.

The data infrastructure discussion reveals how biosecurity and scientific advancement are inextricably linked. Progress in both domains depends on the same fundamental resource: comprehensive, high-quality biological data properly managed and made accessible to legitimate researchers.

Subscribe now

Artificial General Intelligence Forecasting and Scenario Analysis. This report from Sarma, Bhatt, Jacob, and Steratore at the RAND Corporation Center on AI, Security, and Technology (CAST) synthesizes the current state of AGI timeline forecasting across expert surveys, prediction markets, compute-centric models, and trend extrapolation, finding that estimates have compressed toward the 2030s across independent methods while the infrastructure for validating those forecasts remains thin. The report describes a taxonomy of three “cruxes” driving disagreement (scaling sufficiency, diffusion speed, takeoff dynamics) and its argument that forecasts should be treated as scenario-structuring tools rather than point predictions. There’s an interesting disclosure here: the report was primarily drafted by LLMs (GPT-5.1, Gemini 3 Pro, Claude 4.5 Opus) with human direction and fact-checking, and the authors are candid about the recurring citation errors and factual inaccuracies they had to correct in the AI-generated drafts. Maybe the disclosure itself made me extra sensitive to all the LLM tells—it’s not X it’s Y, em dashes everywhere, and tricolons like this one.

The White House released their National Policy Framework on AI. The move that shouldn’t surprise anyone here is federal preemption of state AI laws: the framework explicitly calls for barring states from regulating AI development, which it frames as inherently interstate with national security implications. Beyond preemption, the document covers child safety (parental controls, age-assurance requirements), energy (residential ratepayers shouldn't bear data center electricity costs), IP (the administration believes training on copyrighted material is fair use but says courts should resolve it), and an anti-censorship provision aimed at preventing government from coercing AI providers on content. Notably absent: any new federal regulatory body for AI. The framework explicitly says Congress should not create one, relying instead on existing sector regulators and industry-led standards.

White House National Policy Framework On Artificial Intelligence

234KB ∙ PDF file

Download

Download

Technology Development for Genomics. A new NIH highlighted topic (open through August 2026) signaling investigator-initiated applications are welcome for new methods in sequencing, epigenomics, functional genomics, multiomics, spatial/temporal resolution, direct RNA/DNA sequencing, oligo synthesis, and high-throughput phenotyping of genetic perturbations. Participating ICOs include NHGRI, NIAID, NCI, NHLBI, and others. Not a specific funding opportunity announcement, but a signal of what program officers want to see come through the parent R01/R21 mechanisms. Particularly relevant context given the OMB hold on NIH spending just lifted this week; if new grants are about to start flowing again, knowing what the institutes are hungry for matters.

Posit: Working with Jupyter Notebooks in Positron.

UVA SDS Story of Us: Chapter 4: A Growing School. The latest chapter of UVA’s institutional history of its School of Data Science, covering 2024 to the present. The throughline is a school reaching institutional maturity: first PhD graduates hooded, a retooled master’s curriculum, an inaugural undergraduate class of 75 (now 125), a second building approved for design, and our faculty and staff growing from 30 to over 100. The chapter closes with the death of founding dean Phil Bourne earlier this month. Phil’s legacy is woven throughout, from insisting his office be no larger than any other faculty member’s to spearheading UVA’s Futures Initiative on AI in higher education. When asked what he was most proud of, Phil answered immediately: “Everybody.” A fitting capstone for a school that didn’t exist a decade ago. Phil’s memorial service was live streamed last week, and the recording is available here.

Richard McElreath released lecture recordings and notes from his course, “Statistical Rethinking.” The course focuses on logical and critical statistical workflows, from basic probability theory to causal inference to reliable computation to sensitivity.

Subscribe now

Rahul Garg: Context Anchoring. Part of an ongoing “Reducing Friction” series on martinfowler.com, this one argues that developers cling to long AI coding sessions not because they’re productive but because the decision context lives nowhere else. The proposed fix is a lightweight “feature document” that externalizes the why behind decisions (not just the what), functioning as a living Architecture Decision Record (ADR) that lets you kill a session and cold-start a new one in 30 seconds. If closing your chat window makes you anxious, your context isn’t properly anchored.

Five Things: March 22, 2026. Politicians, security breaches, OpenFold3, the AI “consciousness cluster”, dog’s cancer cured? As always, Matt provides a great recap of things happening in the world of AI/biosecurity and related topics over the past week, and it’s one of the weekly newsletters I look forward to the most.

Charlotte Wickham: Quarto 1.9. The highlights for me: Typst now supports book projects and article layouts with margin content, there’s experimental PDF accessibility compliance for both LaTeX and Typst, and websites can generate llms.txt output so LLMs can read your docs cleanly. More on accessibility via alt text, and automating it with Claude Code:

: The NOFO Graveyard. A former NIH program officer (22 years) responds to NIH Deputy Director Jon Lorsch's blog post defending the agency's reduction in Notices of Funding Opportunities. Ginexi compiled 14 years of NOFO data: NIH averaged ~724 NOFOs per year in 2023–2024, posted 120 in 2025 (an 83% decline, not the mandated 50%), and is on pace for far fewer in 2026. Of 271 opportunities forecasted in 2025, only 120 were ever posted. Lorsch's claim that fewer NOFOs wouldn't mean fewer funded applications is contradicted by NIH's own data showing R01-equivalent awards fell 20% in a single year. Ginexi argues the real administrative burden on program staff isn't NOFO management but politically motivated grant screening, and that centralizing NOFO approval through NIH/OD, HHS, and OMB transfers scientific priority-setting from domain experts to political appointees.

Proteina-Complexa. NVIDIA's new protein binder design framework unifies generative modeling and hallucination-style optimization under one roof, using flow-based latent generation with test-time compute scaling. Over a million binder candidates screened across 133 targets yielding 63.5% hit rates on PDGFR with picomolar affinities and the first-ever de novo designed carbohydrate binders. Self-generated sequences outperformed ProteinMPNN redesign across the board, which, if it holds up broadly, is a shift away from the two-stage generate-then-redesign paradigm.

R Weekly 2026-W13: What’s new in dplyr, Atmospheric simulation.

Simon Willison: Using Git with Coding Agents. A chapter from Willison's growing “Agentic Engineering Patterns” guide. You don’t need to memorize Git's more arcane commands anymore, but you do need to know what's possible so you can ask for it. Good bits on using git log to seed a fresh agent session with recent context, and on how agents turn git bisect from a tool most developers avoid into something you'd reach for casually.

Diego Oliveira, Qian Huang, Teresa Woodruff, and Brian Uzzi: How the 2025 NIH Grant Terminations Varied by Researchers' Demographic Groups. A PNAS brief report documenting who got hit by the 2025 NIH terminations: 2,291 grants, $2.45 billion rescinded, with 52% of allocated funds already spent at the time of cancellation. Early-career investigators and women were disproportionately affected. Among assistant professors, 59.8% of terminated projects were women-led. Women's grants were smaller on average ($940K median vs. $1.4M for men) but had a larger share of unspent funds at cancellation (57.9% vs. 48.2%), meaning more ongoing work was interrupted per dollar. Training and transition awards (F31s, F30s, T34s) were frequently cut. The authors estimate $6.29 billion in unrealized economic output using standard NIH multipliers, though they're careful to frame that as a benchmark rather than a verified loss.

Emily Mullin, WIRED: A Billionaire-Backed Startup Wants to Grow ‘Organ Sacks’ to Replace Animal Testing. R3 Bio, a Bay Area startup backed by Tim Draper and a Singapore longevity fund, is pitching the idea of growing brainless primate (and eventually human) “organ sacks” as replacements for lab animals. The concept: use stem cells and gene editing to grow organized organ structures that lack any brain tissue, making them incapable of sentience or pain. The near-term application is drug toxicity testing in monkeys at a time when US primate supply is constrained after China’s 2020 export ban. The longer-term ambition is growing human organ sacks as a source of transplant organs. The company says it’s currently only working in monkey cells, though a job posting seeks a veterinarian in Puerto Rico to implant embryos in nonhuman primates. The science is plausible according to UC Davis stem cell biologist Paul Knoepfler, but everything about this is still highly theoretical, and a Stanford bioethicist notes the “yuck factor will be strong.”

Alexis Gallagher and Rens Dimmendaal: So Where Are All the AI Apps?. If AI coding tools make developers 2x (or 10x, or 100x) more productive, you should be able to see it in PyPI package creation and update rates. The authors looked, and mostly didn’t see it. Total package creation shows no inflection at ChatGPT’s release. Update frequency for the top 15,000 packages shows a modest secular trend that predates AI tools. The one place a clear >2x effect does appear: popular packages about AI, which jumped post-ChatGPT compared to popular non-AI packages. The authors’ two candidate explanations are that AI-tool builders are the ones most skilled at using AI to build (a skill effect), or that the flood of money into AI is simply paying for more work on AI packages (a funding effect). Probably both, but the data can’t distinguish them. Either way, the Cambrian explosion in all software hasn’t materialized; the measurable effect is concentrated in the AI ecosystem building tools for itself.

and , at : State of AI in the Commonwealth: Teaching & Learning with AI. Part of a rolling series from UVA’s Center for Management IT drawing on nearly 200 sources to map AI’s impact in Virginia. This installment focuses on education, and the most useful data point comes from a McIntire study of 356 students across 6 conditions: the best learning outcomes came from group work paired with a custom RAG tutor, while generic chatbots performed worst. The broader numbers frame the context: the global AI-in-education market jumped 46% in a single year (to $7.57B in 2025), 61% of faculty report using AI in teaching, but 88% of those describe their use as minimal.

And a recommendation: My SDS colleague YY Ahn maintains a public wiki that sits somewhere between a research notebook, a tool log, and a blog. It’s updated frequently and worth browsing if you’re into academic workflows, network science, or the ongoing project of making AI tools actually useful for research. A few recent entries: Claude Scholar plugin, a set of Claude Code skills for academic work (arXiv metadata fetching, BibTeX from DOIs, math verification via SymPy, pre-submission LaTeX checks); How I Use Claude Code, a living document on his terminal/editor/dictation setup; paper notes on topics ranging from LLMs inferring political alignment from conversations to community-centric citation dynamics; and a separate urbanism thread with entries on topics like the true cost of car ownership (read this one!). The whole thing runs on a kind of accumulating-notes-in-public model that I bookmark whenever I find them.

Subscribe now

New papers & preprints:

• Generalist biological artificial intelligence in modeling the language of life

• LLMs Can Infer Political Alignment from Online Conversations

• A harmonized benchmarking framework for implementation-aware evaluation of 46 polygenic risk score tools across binary and continuous phenotypes

• LazySlide: accessible and interoperable whole-slide image analysis

• How LLMs Distort Our Written Language

• ProteinMCP: An Agentic AI Framework for Autonomous Protein Engineering

• LLM agents for biological intelligence across genomics, proteomics, spatial biology, and biomedicine

Subscribe now

Emil Hvitfeldt published a blog post with a Claude Code skill to create alt text in Quarto documents.

I have some personal skills in ~/.claude/skills and I added this one to it. I.e., I added a file at ~/.claude/skills/write-alt-text/SKILL.md with the skill text Emil provides in his blog post.

After writing a Quarto doc, I can just ask Claude Code, “Use the write-alt-text on the file @my-doc.qmd” or I can directly invoke the new slash command available to me with the skill now installed: /write-alt-text @my-doc.qmd.

Here’s what a demo Quarto doc looks like before with some plots using the penguins, mtcars, and diamonds data.

---
title: "Testing alt text"
format: html
---

```{r}
#| message: false
library(ggplot2)
```

## Penguins: Bill Length vs. Flipper Length

```{r}
ggplot(
  penguins,
  aes(x = flipper_len, y = bill_len, color = species)
) +
  geom_point() +
  labs(
    title = "Penguin bill length vs. flipper length",
    x = "Flipper length (mm)",
    y = "Bill length (mm)",
    color = "Species"
  ) +
  theme_minimal()
```

## mtcars: Fuel Efficiency by Cylinders

```{r}
ggplot(mtcars, aes(x = factor(cyl), y = mpg, fill = factor(cyl))) +
  geom_boxplot(show.legend = FALSE) +
  labs(
    title = "Fuel efficiency by number of cylinders",
    x = "Number of cylinders",
    y = "Miles per gallon (MPG)"
  ) +
  theme_minimal()
```

## Diamonds: Cut vs. Price

```{r}
ggplot(diamonds, aes(x = cut, y = price, fill = cut)) +
  geom_violin(show.legend = FALSE) +
  scale_y_log10(labels = scales::dollar) +
  labs(
    title = "Diamond price distribution by cut",
    x = "Cut",
    y = "Price (log scale)"
  ) +
  theme_minimal()
```

After running that new skill, Claude Code confirms what it did:

And my new quarto doc looks like this, now with alt text on each chunk:

---
title: "Testing alt text"
format: html
---

```{r}
#| message: false
library(ggplot2)
```

## Penguins: Bill Length vs. Flipper Length

```{r}
#| fig-alt: |
#|   Scatter chart of penguin bill length (mm) on the y-axis versus flipper
#|   length (mm) on the x-axis, with points colored by species (Adelie,
#|   Chinstrap, Gentoo). The three species form distinct clusters. Gentoo
#|   penguins tend to have the longest flippers but moderate bill lengths,
#|   while Chinstrap penguins have shorter flippers but relatively long bills.
#|   Across all species, longer flippers are generally associated with longer
#|   bills.
ggplot(
  penguins,
  aes(x = flipper_len, y = bill_len, color = species)
) +
  geom_point() +
  labs(
    title = "Penguin bill length vs. flipper length",
    x = "Flipper length (mm)",
    y = "Bill length (mm)",
    color = "Species"
  ) +
  theme_minimal()
```

## mtcars: Fuel Efficiency by Cylinders

```{r}
#| fig-alt: |
#|   Box plot of fuel efficiency (miles per gallon) by number of cylinders
#|   (4, 6, or 8) for cars in the mtcars dataset. Cars with 4 cylinders have
#|   the highest median MPG, followed by 6-cylinder cars, with 8-cylinder cars
#|   showing the lowest and most variable fuel efficiency. The relationship is
#|   clearly monotonically decreasing: more cylinders corresponds to worse fuel
#|   economy.
ggplot(mtcars, aes(x = factor(cyl), y = mpg, fill = factor(cyl))) +
  geom_boxplot(show.legend = FALSE) +
  labs(
    title = "Fuel efficiency by number of cylinders",
    x = "Number of cylinders",
    y = "Miles per gallon (MPG)"
  ) +
  theme_minimal()
```

## Diamonds: Cut vs. Price

```{r}
#| fig-alt: |
#|   Violin plot of diamond price (log scale, shown in dollars) by cut quality
#|   (Fair, Good, Very Good, Premium, Ideal) for the diamonds dataset.
#|   Counterintuitively, higher cut grades do not correspond to higher prices:
#|   Ideal-cut diamonds show a distribution concentrated at lower prices, while
#|   Fair and Premium cuts have heavier tails toward higher prices. This likely
#|   reflects confounding with carat size, as lower-cut diamonds tend to be
#|   larger. All distributions are right-skewed on the log scale.
ggplot(diamonds, aes(x = cut, y = price, fill = cut)) +
  geom_violin(show.legend = FALSE) +
  scale_y_log10(labels = scales::dollar) +
  labs(
    title = "Diamond price distribution by cut",
    x = "Cut",
    y = "Price (log scale)"
  ) +
  theme_minimal()
```

Subscribe now

Workshops for Ukraine

Back in January I gave a 2-hour workshop on R Package Development in Positron, covering package development with devtools, usethis, covr, GitHub actions, pkgdown, writing functions, documentation, exported/installed package data, programming with dplyr, and writing unit tests with testthat and lots of help from Claude via Positron Assistant.

All the materials are online, and the workshop recording is below.

• Rendered pkgdown website: https://stephenturner.github.io/rpkgdemo/

• Full tutorial: https://stephenturner.github.io/rpkgdemo/articles/rpkgdemo

• Source code for the built package: https://github.com/stephenturner/rpkgdemo

Subscribe now

Related posts:1

Share