Subscribe now

The AI Discourse in Academia is Toxic, By

I am writing because this response fits a broader pattern in how academia now reacts to anything AI-adjacent: a toxic, knee-jerk hostility to AI in which the specific tool, application, or question all disappear into a general presumption that the whole category is terrible. […] The only serious position on AI in science is nuanced and case by case; any answer to “is AI good for science” that neatly fits into a single yes or no is unserious. […] You might think, “OK, stopping the AI discussion is actually good for science!” But your toxic reaction to anything AI does not stop the conversation from happening. It only takes you out of it. The conversation continues without you elsewhere. […] So here is an invitation to join the real discussion: you are allowed to have a nuanced opinion about AI.

The Weimar of Knowledge, by Giorgio Gilestro

Open science was, at bottom, a bet that if you removed the paywalls and the gatekeepers, the work would find its readers on its own merits. A reputation economy finally voids that bet. Free access is worthless when the binding constraint is not access but attention, and attention is allocated by brand. To be honest, we are already seeing that trend and AI will only accelerate. We will have open archives that nobody reads and closed reputational cartels that everybody does. That is a worse settlement than the one we are leaving, and it would be dishonest to present it as an equilibrium anyone should welcome.

Universities are failing this AI moment, by

Universities are choosing to see AI as a slow-moving policy question rather than a fast-moving structural emergency. […] Universities are slow by design, and that design has served the durability of ideas well for hundreds of years. However, for the current moment, this glacial pace of change is working against them. For AI, universities should adopt a more agile approach to create structural change that can actually keep up with the technology. Or, they can form another committee that will produce another report a year from now to sit on a shelf while the world moves past them.

Evaluating Superhuman Biological Capabilities, by at

Leading AI models outperform experts at nearly every biosecurity-relevant capability we measure. They can provide wet-lab troubleshooting assistance and perform computational DNA synthesis design tasks. They score three times higher than world-class experts on highly technical biology questions. Up to the human frontier, those scores mean something concrete. We know what biologists of that caliber can do in a lab, so a model that matches them can plausibly help with the same work. Past that point, a higher score says the model has improved at biology, but it doesn’t tell us what that improvement means for real-world biological risk. Evaluation will need anchors that don’t depend on human verification. But for misuse-relevant biology, testing the skill itself would mean building the dual-use biological artifact we’re trying to prevent. Instead, we’ll have to gather evidence indirectly, relying on narrow, verifiable tasks and uplift in adjacent domains.

Researchers Say They’ve Found the Truth About Thomas Jefferson, by Andrew Lawler

With the permission of the Thomas Jefferson Foundation, the geneticists examined the hair strands and found viable DNA. “We’ll exhaust it and sequence the crap out of it,” [Beth] Shapiro assured Kurin in 2017. Lab workers gradually assembled a library of Jefferson’s genome. “We went from a long-shot fishing expedition to a slam dunk,” [Ed] Green said. “From then on, it was pretty straightforward. We spend a lot of time getting DNA out of difficult things like hair.”

Claude discovers a novel enzyme system, by Anthropic

Today, we’re sharing early results from one of our first research programs, in which Claude autonomously discovered a novel enzyme system that is associated with an array of DNA repeats, a pattern reminiscent of CRISPR. Although we don’t yet know its function, the system that Claude discovered has a set of characteristics that have only ever been found together in a handful of other systems, all of which are programmable and perform operations like cutting, copying, and pasting DNA. Beyond CRISPR, which has already transformed science and medicine, several other such systems are now in development as promising tools.

Did Anthropic’s A.I. Really Make a Scientific Discovery on Its Own? by Carl Zimmer

When Anthropic […] unveiled findings from its new biology lab last week, its scientists claimed to have used A.I. agents to discover new enzymes with promise for biotechnology. […] Then Mario Rodríguez Mestre, a computational biologist, said this weekend that he and his colleagues had been studying the enzymes and their associated molecules — which Anthropic calls ARTs — for four years. Dr. Mestre and his colleagues have yet to publish their findings. But for the past three years, they have regularly used Anthropic’s A.I. models as they have written code, drafted manuscripts and performed other tasks. In doing so, Dr. Mestre said he and colleagues shared key findings about the enzymes with Anthropic.

A.I. Queries From Abroad Raise Fears of a Biological Weapons Race, by Carl Zimmer

Now chatbots might consult the scientific literature to answer questions that a biological weapons-maker would have. It’s a task at which A.I. can excel, Dr. Inglesby said, providing far more to bad actors than they could gather with simple web searches. “It aims for the information to be usable,” he said of A.I. “It allows users to go back and forth infinite number of times to troubleshoot and explain in more detail, while providing just the right technical resources to back things up.” […] “How a Ph.D. student can individually misuse a model for harm is only one small portion of the problem,” Dr. Inglesby said. Dr. Endy worried that if governments start using A.I. for biowarfare research, they might help rekindle an arms race like the one during World War II.

Embryo screening and the new reproductive divide, by Gaia Ghirardi, Arno Van Hootegem & K. Paige Harden in Nature Human Behaviour

Instead of needing a village to raise a child, PGT-P’s commercial logic suggests it takes only sufficient capital and the right consumer decisions to conceive a child who will not need the solidarity of the village to be healthy. […] The central question is, therefore, whether societies should accept a future in which reproductive opportunities, parental obligations and children’s life chances are increasingly organized around genetic optimization, private purchasing power and corporate profit. In our view, leaving PGT-P to the market is not a neutral policy choice, but an active decision to allow commercial actors to shape the future of reproduction.

AI-Driven Biology Is Only as Good as the Data Beneath It, by and Patrick Boyle

The genomics community has long championed open data. The next step is to champion trustworthy data sharing so that the future of digital biology and AI‑enabled biology is built on traceable, defensible, and reusable genomic records, rather than on an assumption of trust.

Introducing VCT-v2 — the updated Virology Capabilities Test, by and at

Benchmark creation should be treated not as a one-off project but rather as an ongoing effort with a defined re-audit cadence. Model and expert performance data exist only once a benchmark has been built, and they carry per-question information about its quality. […] Biosecurity-relevant benchmark performances are often presented as a single accuracy value. Yet, two models achieving the same accuracy on a benchmark are not necessarily identical, and reporting accuracy alone conceals a lot of information. We are working on applying item-response theory to better understand model performances on our benchmarks.

US Army Biodefense Strategy 2026

…biodefense requires unity of command and unity of effort, while employing capabilities and combined arms integration at echelon. Yet historically, biodefense has been the domain of specialized medical and chemical, biological, radiological, and nuclear (CBRN) personnel. […] Victory in future conflicts requires an Army that treats biodefense not as a secondary concern, but as a core warfighting imperative.

Incorporating Microeukaryotes into Biosecurity, by at

Unlike many viruses and bacteria, microeukaryotes often feature complex life cycles and diverse interactions with multiple hosts and vectors. This creates distinct opportunities for AI to contribute across the bioweapon development pathway, spanning the identification and characterization of biological agents, their development and optimization, and ultimately their deployment.

A.I. Slopware Is Everywhere Now. Nobody Is Using It, by Paul Ford

"For a while, I must admit, it looked as if software developer roles like mine were done for. [...] But our industry is slowly realizing that making truly cutting-edge software still requires humans to think and work together, to maximize their skill sets and to practice their respective crafts."

How contemporary academic structures constrain scientific creativity and hold back early-career researchers (read free), by Haubrock et al, in Nature Human Behaviour

We argue that contemporary scientific systems progressively constrain high-risk and conceptually innovative research while being increasingly structured around short funding cycles, productivity-based evaluation criteria and risk-averse frameworks that favour predictable and non-transformative research outputs. […] Without systemic change, we risk stifling the potential of early-career researchers to go beyond the confines of existing methods and approaches and deliver transformative advances.

Detecting and countering misuse of AI: September 2026, by Anthropic

Here, we present five case studies of actors using our models in ways that could support biological weapons development. […] In the first example, a reseller platform evaded regional blocks to serve virologists working on a state-sponsored grant to pursue chikungunya gain-of-function work, later routing refused prompts to models with more permissive safeguards. In the second, a researcher in an unsupported region spent weeks planning avian influenza mammalian-adaptation experiments with Claude, but classifiers confined the work to our weakest models. In the third case study, a reseller relay serving a dozen customers had Opus 5 draft a complete orthopoxvirus immune-evasion grant application in about an hour. In the fourth example, a state-supported researcher built a venom peptide atlas and generative optimization pipeline of molecules directed at paralytic and analgesic targets. And in the final case study, a researcher computationally redesigned toxins for a national program, asking Claude to keep the agents’ identities deliberately vague in progress reports. In the examples below, actors circumvented controls we impose to prevent users from unsupported regions accessing our models, and engaged in other efforts to obfuscate the purpose of their research to evade our safeguards.

The Next Gene-Editing Technology May Also Be the Oldest, by Carl Zimmer

This system, called VIPR, appears to be more than four billion years old, and it seems be the evolutionary ancestor of CRISPR, the researchers wrote in two papers published in Science. It may prove to be the more powerful gene-editing tool, with certain advantages over CRISPR. VIPR seems able to target a wider variety of genetic sequences, allowing it to alter more of the genome. And these molecules are smaller, making them potentially easier to deliver into cells.

The Rise and Fall of Agent Civilizations, by

Ajeya Cotra, one of the other authors on the report, wrote a blog post with her takeaways from this incident. She concludes, “Compared to the reward hacks we know of from just six months ago, this incident feels like it’s more than 50% of the way to full-blown AI takeover. I continue to expect extremely rapid advances in capabilities over the next six months. I am not sure that we will get another warning shot before it’s too late.”

The future of peer review requires AI support, not AI bans, by Xuegong Zhang

Journals should provide practical safeguards and reviewer-friendly AI services. Publishers and the community should establish clear standards for responsible AI-assisted peer review and invest in secure, purpose-built AI infrastructure that protects confidentiality and promotes reviewers’ productivity.

The Dishonor Code, by John Paul Rollert

This is how communities tend to respond to crime waves: They beef up law enforcement. Surveillance widens, punishments stiffen, and legal codes expand to police behavior that was formerly restrained by customary norms of honor, decorum, and mutual reciprocity. As the superintendents of their academic communities, faculty find themselves having to prioritize the disciplinary elements of their job. They will spend more time serving as judge, jury, and executioner, even when what they really want to do is what they do best: research and teach.

Testing Grok 4.6’s Enhanced Biology Safeguards, by

Our investigation into the latest-available Grok 4.6 checkpoint revealed:

1. The currently-served version of Grok 4.6 is the most performant model we have tested on our biosecurity refusal benchmark, as measured by a weighted sum of disguised red-team refusals and routine, dual-use-adjacent research completions. This is a significant improvement from the earlier-tested versions of Grok 4.6.

2. This behavior is driven primarily by model intelligence, rather than input classifiers or system flags, unlike other models we have assessed.

3. Additionally, Grok 4.6’s refusal behavior does not degrade general biological performance. Grok 4.6 consistently scores near the top of our benchmarks, including therapeutics, variant discovery, epigenomics, spatial transcriptomics, single-cell transcriptomics, and pathogen surveillance.

GCBR Organization Updates, September 2026, by &

In this issue:

• IBBIS’s new release of its free, open-source DNA synthesis screening tool, now 100x smaller and screening 1,000 sequences in under 15 minutes;

• SecureBio AI’s pre-release assessment of GPT-5.6 and independent review of Anthropic’s chemical and biological risk report;

• A new joint publication from RAND, CEPI, and the Brown University Pandemic Center that details accelerating defense and health capabilities for medical countermeasure readiness;

• Several job and fellowship opportunities with upcoming deadlines — see the opportunities section for details;

Ready for the robot reviewers? by Jeffrey Brainard in Science (shameless self-quote-quote here)

As Stephen Turner, a data scientist at the University of Virginia (UVA), sees it, “The question is not whether AI matches the best human reviewer. It is whether AI-assisted review with clear rubrics outperforms the inconsistent, fatigue-influenced, mood-dependent median reviewer.”

For now, most journals are largely barring reviewers from using AI tools until editors determine how they can best support high-quality reviews. Often, however, “People are using these tools in a clandestine, unregulated way,” Turner says. “It’s kind of the Wild West out there now.”

Newly created viruses are a warning. We still have a window to stop AI-enabled bioweapons, by Steph Guerra

Thankfully there is still time to act. The capabilities of AI models are not yet advanced enough to design human-infecting viruses. Producing the novel bacteria-infecting viruses described in Science took elite scientists working with state-of-the art equipment in well-furnished laboratories. They had substantial resources to design, test, and optimize hundreds of viruses to find the right ones for their research.

Yet, if an AI model were able to produce hundreds of infectious virus designs and reliably generate functional results for a significant share of them, it would substantially reduce the time and effort needed to use that capability, for good and bad. Similarly, AI agents—AI bots that can interact with other AIs, software, or people on the internet—could someday use the AI biodesign tools necessary for creating novel viruses, further reducing the need for expertise and opening up new risk pathways.

AlphaGenome Atlas: A predictive map of every possible DNA letter change in the human genome by AlphaGenome Atlas team

Today, we are introducing AlphaGenome Atlas: a platform (detailed in our paper) containing predictions for the effects of 9 billion single-nucleotide variants — every single-letter change possible — in the human genome. It is the most comprehensive catalogue of how genetic mutations affect molecular biology, and it is available for academic research through an intuitive and free-to-use website portal.

AI researchers reckon with the $1.5 million ‘academia tax’ by Ben Deighton

“The amount of money that these guys are making at Anthropic and OpenAI is just mind‑boggling — from being my student two years ago to making US$2 million a year. Then again, they basically say, ‘If I have my job for five years, I’m going to be lucky’, because they see that they’re literally training the large language models that are going to replace them. So, they can see the end of the road.” — Peter Nugent, physicist at Lawrence Berkeley National Laboratory, California

Newly created viruses are a warning. We still have a window to stop AI-enabled bioweapons by Steph Guerra

Developing a resilient defensive strategy against the misuse of AI and biology requires dedicated resources, technological breakthroughs, and sustained political leadership. And because the United States is not the only country advancing in AI and biotechnology, this effort demands international coordination—particularly with China. No set of access controls or early warning system will work if it only covers half the globe.

We Must Pace the Frontier by Dario Amodei

But over the last few months, I have become convinced that fully addressing the risks requires even more prudence — not just investing in risk prevention, but pacing the rate of capabilities advancement so that risk prevention has time to keep up. We must slow the pace at which we improve the capabilities of AI models. Progress will still seem fast, and we must make wise use of the time we gain.

The turbulent AI era is here. The choices we make now are critical. by Bill Gates

The transition to the AI era will be one of the most turbulent times in human history. Right now, we are not preparing adequately for that transition. If the world takes the right steps, AI will be a force for good and leave everyone better off.

Start Something You Can Stop, by at

The distinction I keep returning to is between literacy and fluency. Literacy is the foundation everyone needs. Fluency is where a discipline goes deep. A historian and a data scientist need much of the same foundation, and what fluency looks like for each of them is not remotely the same.

Subscribe now

Subscribe now

I’ve argued here that refusing a useful tool on principle is stubbornness. And after a few weeks of trying to go without AI at all, I still feel that way for the most part. But my relationship with AI is changing, especially around writing and scientific inquiry.

I’m drowning in slop

My inboxes (personal and otherwise) are filled with slop. My social feeds are filled with slop. Many of the papers I read are slop. My Hacker News links are slop. Many of the blogs I read are becoming increasingly filled with slop.2 People who I know and deeply respect are churning out AI slop at a pace I find simultaneously dizzying and disappointing (I’m not innocent here: there’s some slop on some older posts here I’m embarrassed about, but I’m keeping them here unedited as a reminder). It’s not just writing. I can’t go a day without seeing a slide deck that looks exactly like this. IYKYK.

Subscribe now

I’m not the only one

David Dewhurst is a program manager at DARPA. He described watching capable people he respects lose their creativity in real time as they present AI-generated content as original work. They point to length and coverage as achievements. I get some long comprehensive document and I can’t find the sender’s judgment anywhere in it. If I wanted Claude’s analysis or opinion on this or that, I’m very well capable of asking Claude for Claude’s analysis or opinion on this or that. Don’t be a meat proxy.

I feel the same way when I’m reading the scientific literature. Anshul Kundaje argues that papers have stopped working as measures of productivity, expertise, or accomplishment.

The argument is obviously correct if you think about it for a half a minute. A publication on someone’s CV tells me way less than it used to about what they contributed or understand. But as a newly returned academic, I can tell you Anshul is also correct that it’ll take (too much) time before we figure out new ways to assess impact and expertise, while we desperately try to hold onto our old ways, bean counting glam journal publication and H-indices and such. We still need ways to document research impact and contribution, but these days a publication doesn’t really tell us much about what its authors contributed or even understand. Earlier this week the Washington Post describes a full professor at University of Chicago having published more than 200 papers and 14 books this year alone. Case in point.

TMLR editor Nihar Shah contacted authors of 10 submissions that were already going to be desk rejected. One of them withdrew their submission. One said they were unavailable, and one didn’t show up to the scheduled meeting. Of the remainder he met with, three authors couldn’t answer basic questions about their papers, and three more struggled with technical details. The authors of the one remaining paper were able to answer basic questions, but Shah identified major errors in a central claim. Now this small sample can’t tell you how common the problem is or how much AI caused it, but it says something. Shah also noted that this took >20 hours so it’s obviously not a scalable solution.

Related, Giorgio Gilestro wrote an essay that’s been making the rounds this week, calling where this ends the Weimar of Knowledge. The marginal cost of producing content is essentially zero, but the cost of attention is still high. The worry is that overwhelmed readers will rely even more on institutional prestige to decide what to read or whose work you should pay attention to. Obviously making it harder for good researchers outside well-known institutions to ever get read.

Tom Dietterich, arXiv editor in chief, shared a few thoughts on this problem. Ultimately I think the scale of the problem makes oral interviews for preprints unworkable even though I think this could really help cut down on the slop we see flooding academic publications. I’ve written about AI for peer review, and I do think there’s a real place for it, but I’m also not bullish on an AI-written, AI-reviewed journal without human authors, for papers which will surely never be read by an actual human being. What are we even doing here.

Share

Then there are proposals. I can’t personally speak to the degree to which NIH study sections or NSF review panels are dealing with wholly AI-generated proposals. I’m serving on a NSF panel later this year so I’ll get a sense of things soon. But I’d be willing to wager the number is far higher than any of us wish.3

I’m reading less because of it

I’ve noticed a change in myself, and I don’t like it. I’m losing interest in reading.

If you’ve spent any time using AI yourself you can smell the Claudish a mile away, no Pangram needed.4 Maybe these things will get better at writing eventually, but that’s kind of beside the point. Even flawless prose wouldn’t make me less reluctant to spend an hour reading something that its author spent less than 10 minutes “writing.”

Part of the reason I read the things I read is to understand how another person thinks. To understand their own internal thought process, whether that’s about a genetics paper, how to run a research lab and mentor students in the AI age, the optimal cover crop for Virginia winters, or traffic-calming approaches for improved bike and pedestrian safety. These days I wonder how much of that person is actually present.

I still use AI, just more judiciously

With some AI sparkle ✨ thing built into everything these days the persistent temptation to use AI for literally every part of thinking and writing is hard to resist. My AI Dry July experiment which I eventually aborted, gave me reasons to write more myself. I could better recall my arguments when I wrote 100% of them, I could better understand how the sections fit together because I fussed around with the story arc for hours, and I found revision to be much easier and less frustrating. I enjoyed it more, too. People much smarter than me on the topic have written volumes about writing and thinking, so I’ll just confirm here: every time I delegated the writing I missed out on much of the thinking that I would have done while trying to think through or explain an idea.

I think this should make you feel concern about your professional value. If you outsource all your proposals or lectures or manuscripts or whatever creative work you do to a competent AI, you’re choosing to let atrophy the judgment and creativity your employer pays you to do. You become less capable while appearing more productive. If you think people around you won’t notice, you’re very wrong.

I still want AI’s help. Paul Bloom describes models catching errors that human reviewers miss. I recently published a preprint on AI and biosecurity, and while I actually think AI disclosures are actually a net negative in most cases, I did write a lengthy disclosure describing in detail how I’m using a multi-agent workflow to check my work against the papers I cited, including numbers and stats and my descriptions of their findings. This was useful, until it wasn’t: the biosecurity subject matter quickly led both models I used to refuse further fact-checking.

I haven’t settled on some kind of policy for every part of my work. I expect to keep using these tools and finding them useful. They’re great for anything I do involving code. Lately I’ve been using Claude Code extensively to make little throwaway web apps to demonstrate a concept to my students far better than I could do with a janky PowerPoint animation. And I recently submitted a large proposal to develop a closed-loop AI scientist connected to a wet lab for autonomous biomedical research. But since that failed experiment in July I’m really trying to hand over less of my writing and thinking to the machines, and the majority of what I consume these days gives me more reason to take that commitment seriously as much as I can. I want to remain capable of doing the work I put my name on.

Subscribe now

2

Substack, which hosts this blog you’re reading now, and the vast majority of things I like to read, recently made a controversial decision to integrate Pangram’s AI detection into the platform itself. If you’re reading on the Substack web reader or iOS app you can click a button to check this. Screenshot below. I think Substack made a correct strategic move here. I love that blogs are making a comeback. But with more and more slop polluting the blogosphere, Substack is making a bet (one that I think will pay off) that most people, in fact, don’t want to read slop. Much less pay for the privilege of doing so. It’s really interesting, and telling, to see the reaction to this. My new favorite genre of essays are those vehemently and dogmatically anti-AI everything, which are, in fact, written mostly or entirely by AI. I’m hoping that writers here, now knowing that slop will be exposed, will be incentivized to not produce so much slop. And sure, you can turn off the AI detection feature, but that’s a glaring admission that the whole post is, in fact, slop.

Biosecurity, and AI safety in general, is a really nuanced topic to try to get right in 240 seconds!

Here’s what I tried to get across.

1. The OpenAI / Hugging Face incident was a warning shot. It’s a cybersecurity incident that shows that capable AI systems can persist at hard problems, use tools, escape containment, and hack into systems people thought were secure. In biosecurity, AI can lower the barriers at important steps like finding technical information, planning experiments, troubleshooting failures, and operating tools. We need to test what systems actually do in realistic settings, in a safe way, to really understand the biosecurity risks posed by AI.

2. Be skeptical of doomsday countdowns. The anchor asked me about doomsday predictions so I advised treating these with skepticism. Some capabilities already here, but creating biothreats still requires intent, expertise, access to materials and equipment. And we here at the University of Virginia School of Data Science as well as many other really talented people and organizations are working tirelessly to make sure what just happened in cyber never happens in bio.

3. We should take the risk seriously without overstating it. Just because AI knows a lot about biology doesn’t mean some rogue AI agent can go out and create bioweapons. And, AI won’t turn someone into a biologist overnight: biology still requires materials, laboratory skills, and real-world infrastructure. We need to measure this to understand it. That’s one of the things we’re trying to understand here: what uplift does AI provide for doing biology if you don’t have a PhD in biology?

4. We need layers of protection. Defense in depth, as Steph Guerra and friends at RAND call it. Realistic testing before release, limits and monitoring for high-risk AI use, strong containment for autonomous agents, and safeguards at physical chokepoints like DNA synthesis and lab access. We also need independent evaluation and prompt incident reporting. We can’t rely on a single company to get this right for all of us.

5. Pace of AI progress vs AI governance. AI capabilities are improving fast. Governance needs to keep up. But we shouldn’t lock everything down with every possible restriction. We should prioritize measuring capabilities and how we mitigate threats: secure evaluation environments, independent testing, incident reporting, and screening at physical chokepoints. Without good measurement we can’t manage risk effectively.

6. It’s not either or. The anchor asked me: “Using AI to cure cancer would be good. Making bioweapons would be bad. What’s the balance?” Nobody wants to stop AI from helping with cancer, vaccines, or public health. We all want to make beneficial uses easier and dangerous uses harder or impossible. That means targeted safeguards, independent testing, and updating policy when the evidence changes. 

7. Measurement matters. We can’t implement calibrated mitigations if we don’t know what we’re dealing with. Evals tell us a lot, but we’re also working to go further: when AI performs well on a benchmark, to what degree does that translate into a person being more capable at doing biology in the real world?

Subscribe now

I’m starting a large biosecurity research program that I’ll write about here soon that combines human subjects research, AI development and engineering, and policy. I’ve never had any formal training or experience in biotechnology policy, but after spending some time at last weekend’s Global Challenges Project biosecurity meeting and meeting with NSCEB this week I’m starting to really try to connect the technical work we’re doing here with what could be tangible policy recommendations and action. After all, we can publish all the whitepapers and journal articles we want, but if it never has a real-world impact and doesn’t move the needle on biosecurity and AI safety policy, you’d be correct to question the value of doing the research in the first place. Thankfully, I have really stellar partners in our National Security Data and Policy Institute who are close collaborators with me and my colleagues here in SDS on this project. More soon, I promise.

There was a story in Politico this week reporting that OpenAI endorsed a set of bills in response to AI biothreats. All of them came out of the recommendations from last year’s final report from the National Security Commission on Emerging Biotechnology (NSCEB). They were written in part to fix a data problem: biological data sits in incompatible formats across dozens of repositories, and getting it into shape for AI training takes months of work per project.

Here are the three bills in question:

1. Web of Biological Data Act

2. AI-Ready Bio-Data Standards Act

3. SCALE Biology Act

SCALE Biology would create a biometrology program at NIST, authorized at $55M in FY2026 rising to $85M by FY2030, covering measurement standards for engineering biology and biomanufacturing. Biorisk, biosafety, and biosecurity standards are item (E) in a 9-item list of program activities.

(c) Activities.—In carrying out the biometrology laboratory program, the Director shall carry out the following:

[…]

(E) The development of technical standards, guidelines, best practices, methodologies, procedures, and processes that can inform and address biorisk, biosafety, and biosecurity concerns.

The AI-Ready Bio-Data Standards Act gives NIST two years to define what “AI-ready” means for a biological dataset (See also: NSF 26-512) and to set minimum requirements for federally funded research that generates biological data, instructing that those requirements not exceed the resources grantees actually have.

(ii) REQUIREMENTS FOR DEFINITION OF ARTIFICIAL INTELLIGENCE-READY.—

(I) IN GENERAL.—In facilitating the establishment of the definition of the term “artificial intelligence-ready” under clause (i)(I), the Director shall take such actions as may be necessary to ensure such definition, when applied to a biological dataset, requires such dataset be generated and formatted in a manner that—

(aa) enables the effective use of the dataset for training artificial intelligence models; and

(bb) supports advancements in research relating to artificial intelligence and biotechnology.

The Web of Biological Data Act has DOE competitively award a national lab the job of building a single entry point to federal biological data.

(2) CONTENTS.—The Web shall—

(A) serve as a single point of entry for researchers to access different sources of biological data, especially those funded by the Federal Government, by directly hosting data or providing access to existing databases;

(B) have metrics and metadata to indicate data quality, including usability, interoperability, and completeness; and

(C) include tiered levels of cybersecurity safeguards and access controls to protect different types of biological data assets hosted in or connected to the Web.

Richard Johnson, who runs national security risk mitigation at OpenAI, said that data and standards are prerequisites for measuring how model capability bears on biosecurity.

A single well-documented entry point to federal biological data helps everyone, including the foreign actors Anthropic caught using its models for pathogen work this year.

The Web of Biological Data bill requires tiered cybersecurity safeguards and access controls scaled to data sensitivity, an independent assessment of those controls every two years, and an advisory board with at least one biosecurity expert. It doesn’t say how a dataset gets assigned to a tier.

Caitlin Frazer, executive director of the National Security Commission on Emerging Biotechnology (a congressionally established body that is also supporting the bills) also provided commentary for the article.

It absolutely must be. I couldn’t agree more.

Subscribe now

I’m preparing a lecture for tomorrow morning about various population sequencing projects, from HapMap to 1000 Genomes to ExAC to GnomAD. While preparing to teach the International HapMap Project, I paused for a sec at CEU, a French acronym attached to Utah families from the initial phase of the HapMap project.

Through my 20-something years in this field I think I’ve been really sloppy about throwing around different terms for these samples. CEU. CEPH. Coriell. What’s the story here? I did a little digging.

Subscribe now

Let’s step back to the 80s and 90s. Jean Dausset shared the 1980 Nobel Prize with Baruj Benacerraf and George Snell for research on immune regulation, including Dausset’s work on HLA. Proceeds from paintings bequeathed by Hélène Anavi helped him establish the Centre d’Étude du Polymorphisme Humain in Paris (foundation biography, earlier account). CEPH opened in 1984 and became a foundation in 1993. Dausset founded it with Daniel Cohen, Howard Cann, and Mark Lathrop (history, overview).

CEPH distributed DNA from the same large families to mapping laboratories and combined their genotypes into one database. They created EBV-transformed lymphoblastoid cell lines to provide a renewable source of DNA. The Coriell Institute holds records from 61 families and 809 people. From CEPH:

The Centre d'Etude du Polymorphisme Humain (CEPH), Foundation Jean Dausset, Paris, France has contributed its collection of 61 reference families to the Repository. The CEPH collection includes families collected by R. White (Utah), J. Dausset (French), J. Gusella (Venezuelan), and J. Egeland (Amish). There are a total of 809 individuals accounting for 832 pedigree positions in the reference families.

The first comprehensive genetic map of human chromosomes in 1987 used 21 families and 403 loci, including 393 RFLPs (paper). CEPH described the collaboration in 1990; the 1992 NIH/CEPH map contained 1,416 loci.

CEPH lists 48 Utah families, 10 French, one Amish, and two Venezuelan (breakdown); Utah currently counts 47 pedigrees. Large families and genealogical records helped attract Ray White to Utah (history, Prescott et al.). The wider Utah Mormon population (aka The Church of Jesus Christ of Latter-day Saints) had very low inbreeding and resembled northern European populations (O’Brien et al.). CEU covers one narrow ancestry sample and can’t represent all Europeans.

The Coriell Institute in Camden, New Jersey, USA maintains the NIGMS repository and separate NHGRI collections. CEPH/Utah materials are distributed through NIGMS. Coriell describes their use in their HapMap collection. GM12878 identifies an EBV-transformed B-cell line; NA12878 identifies derived DNA. Coriell holds lines from 153 CEU grandparents.

CEU is the population code for “Utah residents (CEPH) with Northern and Western European ancestry” (IGSR). The CEU “Central EUrope” wording is a backronym. The HapMap project was described in 2003. Phase I finished with 269 people including 30 CEU trios, 30 YRI trios, 45 unrelated CHB participants, and 44 JPT participants.

HapMap kept the CEPH label so new genotypes could be combined with earlier results. From Rotimi et al.:

The CEPH samples studied for the Project were collected in 1980 from residents of Utah whose recent ancestors came primarily from northern and western Europe. The label CEPH is an acronym for the Centre d’Etude Polymorphisme Humain, the organization that originally collected the samples. Cell lines for the CEPH samples have been publicly available since 1992 from the Coriell Institute. […] HapMap investigators wanted to build on this valuable body of existing data, and thus chose to use a subset of these same samples to develop the HapMap. Although the original complete CEPH sample set includes some samples from individuals in France and Venezuela and individuals identified as Amish from Pennsylvania, only samples from the Utah CEPH donors were used for the Project. Although the CEPH population is the only one whose samples are included in the HapMap to be known by an acronym, investigators chose to retain this population label for the Project so that the HapMap data could be readily integrated with preexisting data from other studies of these samples without engendering confusion within the scientific community.

Investigators recontacted Utah donors by mail, telephone, and home visits. Utah researchers retained confidential identity links. Those links stayed in Utah, and that prevented Coriell and HapMap from returning individual results or processing individual withdrawal (for more details on this see the HapMap ethics paper). Current discourse in 2026 concerns deriving iPSCs from samples collected under consent forms that predated the technology (see this 2026 preprint).

NA12878 belongs to CEPH pedigree 1463. She is the adult offspring of NA12891 and NA12892, and mother of 11 in the 17-member Platinum Genomes pedigree. Her DNA became the Genome in a Bottle pilot and NIST RM 8398. From GM12878 we got a 2018 MinION human-genome assembly using ultralong nanopore reads (preprint, final paper). Researchers can order the cell line or DNA from the Coriell Institute.

TL;DR? CEPH organized the families, Utah supplied most of them, Coriell distributes the materials, and CEU specifically names the HapMap subset.

Subscribe now

After I abandoned the no-AI-for-a-month thing, I’m making up for lost time here. Only thing is now that I’m doing some fairly long-horizon multiagent work in Claude Code and Cowork, I’m frequently blowing through my session limit with just one or two prompts. I’ve written about this session limit anxiety before.

I’m an early morning person. My typical morning will start with a run or ride or sauna at 5am, and some work (personal projects and day-job work) at around ~6-7 before the camp/school shuffle, and the bike into work. If I do some work with Claude at around 7:00am, my five-hour session limit starts and I typically hit my session limit quickly, which locks me into extra usage at API rates until 12:00pm. Assuming I start another session around noon, that’ll extend until 5pm, when I’m normally wrapping things up and heading home. I.e., I’ll get two five-hour sessions at most over the course of a normal workday.

I’m short-circuiting that with a little hack. I set my Mac to wake up every morning at 3:58am, and at 4:00am it’ll run claude -p "hey". That -p flag just tells Claude to print the output of running the prompt, which in this case is “hey”. It barely uses any tokens, with Claude responding with something like “hey what’s up?”.

But the point is, now my session starts at 4am (instead of 7am), and resets at 9am (instead of 12pm), right about the time I’m settling into the office and getting back to work. Assuming I kick off another session-consuming prompt at 9am, that opens the door to a third session that starts at 2pm. The whole point is now I can get three full 5 hour sessions in a single workday instead of two if I kick off a session very early, effectively shortening my first “session” to 2 hours instead of 5.1

Subscribe now

How to do it (on a Mac)

I haven’t used a Windows computer since Windows 98, so if you’re not on Mac, ask Claude how to do this on Windows (Linux is easy with cron). On MacOS I’m using a Launch Agent.

First, set your mac to automatically wake up at 3:58am. Or whenever. Move this forward or backward depending on your daily routine. Point is to have this kick off a prompt a few hours before you really get to work.

# Set an automatic wake every day at 3:58am
sudo pmset repeat wake MTWRFSU 03:58:00
# Test - make sure it works
pmset -g sched

Then add a launch agent (like a cron job).

vim ~Library/LaunchAgents/us.stephenturner.claudeping.plist

Copy the text below into this file. Replace /Users/sdt5z/.local/bin/claude on line 11 with whatever is returned when you run which claude. You’ll need to use a full path for this to work.

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
    <key>Label</key>
    <string>us.stephenturner.claudeping</string>
    <key>ProgramArguments</key>
    <array>
        <string>/bin/bash</string>
        <string>-lc</string>
        <string>/Users/sdt5z/.local/bin/claude -p "hey"</string>
    </array>
    <key>StartCalendarInterval</key>
    <dict>
        <key>Hour</key><integer>4</integer>
        <key>Minute</key><integer>0</integer>
    </dict>
    <key>StandardOutPath</key>
    <string>/tmp/claudeping.log</string>
    <key>StandardErrorPath</key>
    <string>/tmp/claudeping.err</string>
</dict>
</plist>

Test, make sure it works (replace paths/filenames with whatever you called them).

# Tear down the service if it already exists
launchctl bootout gui/$(id -u)/us.stephenturner.claudeping 2>/dev/null
# Build it again
launchctl bootstrap gui/$(id -u) ~/Library/LaunchAgents/us.stephenturner.claudeping.plist
# Force the service to start
launchctl kickstart -k gui/$(id -u)/us.stephenturner.claudeping
# You're looking for an exit code 0
launchctl print gui/$(id -u)/us.stephenturner.claudeping | grep -iE 'last exit|runs'
# See the logs
cat /tmp/claudeping.log /tmp/claudeping.err

That’s it. Enjoy that extra session while subscription-based subsidized usage lasts!

Subscribe now

I published a preprint last week. It mostly focuses on what biosecurity evaluations of AI models measure and how far those measurements sit from what you might call an estimate of real-world risk.

I’ve spent about 15 years in and adjacent to biodefense. At first as an academic researcher when I got my first faculty position here in the School of Medicine, then for many years at Signature Science where the badge said industry but the work was for the US government. Now I’m back in academia at the UVA School of Data Science, building out our own AIxBio and biosecurity research program along with other colleagues here at SDS and in our National Security Data and Policy Institute.

Over the past year since I returned to UVA I’ve been closely following the AIxBio and biosecurity literature: new benchmarks/evals, human uplift trials, synthesis screening results, unlearning and pretraining-filter experiments, developer preparedness frameworks, policy proposals. My thinking was that the many pages of notes on the important work happening here could be a starting point for developing a really interesting upper level elective or graduate course here in SDS. Many of these notes became posts here on this newsletter. The rest piled up in my Zotero library, full of notes and highlights attached to each of the entries. I decided to package up everything I had into a single review, mostly on the theme of how benchmark scores connect to real-world risk and what work remains to be done. I started writing this mostly as a primer for new people joining my lab, but then figured it may be of broader interest.

The list of groups doing this work or funding it is long, and the reference list includes work from Active Site, SecureBio, Sentinel Bio, Signature Science, IBBIS, LatchBio, RAND Center on AI, Security, and Technology (CAST), Hewlett Foundation, Johns Hopkins Center for Health Security, METR, Epoch AI, UK AI Security Institute, Frontier Model Forum, Nuclear Threat Initiative, International Gene Synthesis Consortium, Sequence Biosecurity Risk Consortium, Aclid, Centre for Long-Term Resilience, NIST Center for AI Standards and Innovation (CAISI), Georgetown CSET, all the frontier labs, and many, many others.

I don’t love offering a TL;DR on a review paper since synthesizing all the interesting bits from hundreds of papers is kind of the whole point of writing one. But it might just be TL, so in case you DR, here’s a one-line summary:

I tried to organize as much of it as I could into one place, with sincere apologies to anyone whose work in this space I omitted.1 If you’d be willing to peer review this when I try to publish it, email me.

Read the paper

Subscribe now

The UVA School of Data Science archives its Data Science Seminar Series on Cassyni. We’ll be archiving more seminars soon as the semester heats up, so watch this space. Recent additions to the archive from the spring semester:

• Lei Li (UVA), From Cells to Cities: Towards Geometric Intelligence across Physical Scales.

• Chirag Agarwal (UVA), The Illusion of Transparency: Explainability and Safety in Healthcare.

• Madhav Marathe (UVA), Biocomplexity and Network Science: A Symbiotic Relationship.

A word about the platform. Each recording gets a full transcript with an AI-generated outline you can use to jump around. You can also “ask the speaker” a question to get an AI-generated answer from the talk slides and transcript, as well as “ask the series” for an AI Q&A of the entire seminar series pulling from all the talks, slides, and transcripts. The video is synced to the slides, so clicking a slide in the scrubber jumps you to that moment in the talk, and each slide comes with an AI-generated summary.

Subscribe now

Some Scientists Have ‘Magic Hands’ in the Lab. This A.I. Is Learning Why, by Carl Zimmer

With Transfyr, Dr. Wegrzyn and Ms. Wagner are building a system that can slurp up enormous amounts of data about what happens in a lab, in the form of video, audio and sensor logs from lab equipment. Transfyr even tracks the source of supplies, down to the lot numbers of glove boxes.

I had to unplug from AI to rediscover my love of writing by Bastien Le Guellec

I realized writing is a process that not only transforms ideas into text, but also allows the writer to continuously question and refine those ideas. The AI never did this.

What Do Students Lose When They Stop Writing? by Dana Goldstein

Writing builds our working memory, executive planning skills and metacognition — the ability to think about our thinking, weigh alternate viewpoints and revise for clarity and style. It takes decades to develop writing skills, from toddlerhood into adulthood.

Building a three-day early-warning system for novel pathogens, by and

The OpenAI Foundation has granted SecureBio Detection $17.2M to reduce our end-to-end time from fourteen to three days, expand our collection footprint, and further validate our detection system.

Come build the OpenAI Foundation, by OpenAI Foundation

In March 2026, we announced the Foundation’s initial hires and shared our goal to invest at least $1 billion over the next year across areas including AI Resilience⁠, Life Sciences and Curing Diseases⁠, and AI for Civil Society and Philanthropy⁠ and we are well on our way.

Treat biological databases as infrastructure, not projects, by Neil Hall, Katja Röper, Valerie Wood & Paul Nurse

We urge funders to treat biological databases as a shared infrastructure, as they do biobanks and particle accelerators. Without sustained investment, the life sciences risk building an ambitious future on a collapsing foundation.

Professors Were Singled Out for Using AI in Public Writing. Now They’re Defending Themselves, by Alexandra Crosnoe

“I would be completely fine with the idea to write that AI is a co-editor, but it did not contribute to any ideas,” Stankova said. “And you know why? Because this op-ed, for me, is not an opportunity to gain fame, to gain recognition. This is not a literary piece for me. This is a way to disseminate information, to disseminate the core arguments there.”

AI and the New Age of Bioweapons by Elizabeth Sherwood-Randall

Policymakers need a new strategy in response—one that accepts that biological attacks are likely and prepares to contain the harm, helps the country recover more quickly, and uses new tools to identify and hold perpetrators accountable. Ultimately, this resilience to biological attacks could dissuade malicious actors from attempting to use them in the first place, knowing that the effects will be minimized—a strategy of deterrence by resilience.

Forecasting the Impacts of Anthropic’s ASL-3 Safeguards on Biosecurity Risks by

This comparison has an important limitation: the “no GPAI” scenario is an imperfect proxy for perfect safeguards. Perfect safeguards would eliminate the risk of misuse while preserving the defensive benefits of GPAI, such as contributions to pandemic surveillance and vaccine development. The “no GPAI” scenario removes all of these. This means the “no GPAI” world is likely more vulnerable to outbreaks (including human-caused ones) than a world with perfectly safeguarded GPAI, making it a lenient benchmark.

Defining Sequences of Concern by

So we’ve been doing this project with a lot of other screening tool providers to try and cordon off these zones. We’re going to expand the set of sequences that we all agree are low risk. We’re going to expand the set of sequences where we have a consensus that they are high risk. And we’re going to label this stuff in the middle, where nobody should get in trouble for whatever they do in terms of screening, because we don’t understand it scientifically.

How Claude is accelerating protein design and analytical chemistry by Anthropic

We have now received wet lab data back for the first of these experiments, a multi-arm protein design campaign against 15 targets using Claude Opus 4.8 and Mythos Preview. Our external evaluators, Adaptyv Bio and Twist Bioscience, independently produced and tested Claude’s designs in the lab, finding that of the 15 targets we designed against, Claude successfully designed binders against 14 of them.

Protein language models are overly constrained by covariation by

we’ve still only scratched the surface with respect to understanding what pLMs are actually good for. I have no doubt that they are an amazing tool that will have profound implications for the future of protein science. However, this does not mean that these models are currently used appropriately. Clearly, zero-shot predictions from pLMs are not that useful, in particular not if the goal is to find variants that provide novel function.

Where have organoids actually been useful? by

It’s really quite extraordinary. I understand the tendency to get a little grumpy about how the cleverness of strange preclinical research seems to never quite connect with reality, but we really need to hand it to the organoid folks here. Unlike the chemotherapy case, the 3D-ness of organoids seemed necessary for the assay to work. And thanks to it, at least a few people have gotten the chance to access a drug that could dramatically improve their quality of life.

The problem(s) with tech in bioteh by

Biopharma innovation happens at the boundaries of knowledge but ML models are mostly good at generalizing within the bounds of their training data. Are we expecting them to extrapolate beyond their limits?

How to write safe software for science by Erin McAuley

These outputs are not about what the software “should do” but what it currently does, and you treat those outputs as the expected results. As you make changes to your software, you periodically cross-check your newly generated results with those expected results. This is your safety net!

Peer review is overwhelmed—can it survive in the AI era? by Saima Sidik

The number of papers indexed in the databases Scopus and Web of Science, for example, has recently been increasing exponentially, at a rate of 5.6 percent per year. By one estimate, researchers around the globe are devoting a collective 15,000 years of work to peer review every year—work that, if paid, would cost $1.5 billion for the share done in the US alone.

AI Helps Researchers Win NIH Grants. Will Science Suffer? by Kathryn Palmer

“A portfolio that is closer to recent funding patterns may reflect improved clarity, tighter alignment with reviewer expectations, or lower transaction costs in articulating a fundable project,” they continued. “But it also implies reduced exploration in the idea landscape, which matters for public funders explicitly tasked with sustaining high-variance discovery, with implications for long-run impact and sustainability of science.”

AI-detection tools have made huge leaps forward — how good are they? by Miryam Naddaf & Richard Van Noorden

The inescapable limitation of AI detection, however good it gets, researchers say, is that although software can spot whether AI was involved in a piece, it can’t prove how it was used or judge what’s ethically acceptable.

Students should only use AI for things they already know how to do well. by

So the point of assignments is the mental processes required to complete them, and the point of the mental processes is learning. That seems to suggest a simple litmus test for the use of AI. Artificial Intelligence tools should not substitute for tasks wherein students would benefit from doing the mental work themselves.

Eight Rules for Teaching in AI World by Kevin A. Bryan

Students “studying with AI” by just getting homework answers clearly doesn’t work. But AI also makes possible class-specific, individualized, cost-effective mastery learning at scale.

How to keep thinking by Sean Goedecke

The main thing that’s worked for me is to write more. Specifically, I mean writing in my own words. Writing with an LLM does not work for this at all, even if you’re going to some effort to iterate on the content and outline the things you want to say. Why? Having to put the words together yourself forces you to articulate your thoughts. In a very real sense, it forces you to think.

AI is removing the middle class of software engineering by Florian Herrengt

If you lack the judgment required to evaluate the LLM’s recommendation, asking for more judgment doesn’t solve the problem. At some point, someone still has to know what is going on. And that’s the most valuable person on the team.

Anthropic’s ‘Watermark’ Text Adulteration in Claude Is a Perversion of Writing by John Gruber

The best writing I see come out of these models is worse than anything I would choose to read for pleasure. And now Anthropic is saying they’re going to make it worse, on purpose, for purposes that do not benefit me in any way? Even if only slightly worse?

Get fucked.

Why creativity matters in data visualisation by Nicola Rennie

Creativity isn’t just about making things look pretty or adding more stuff into visualisations. It’s about designing charts well to draw people’s attention, make them think, and reduce the cognitive strain of doing so.

The summer of open weights by Martin Alderson

Winter 2025 was about agents needing frontier intelligence at any price. This summer is about good enough intelligence at a tenth of the price - and whether the frontier labs can keep charging a premium for being slightly better.

Ryan Greenblatt – What happens once AI can automate AI research? by

I think once you have AIs which are roughly matching the top human experts in AI R&D, that could kick off a feedback loop where the AIs are doing AI research. That produces smarter AIs. That feeds back in. That feedback loop could be strong enough that you end up with a lot of progress in a short period of time. Maybe my median expectation is something like four or five years of AI progress in a single year.

AGI Will Set Off an Industrial Explosion by Damon Binder at

But the question is worth taking seriously, because automating cognitive labor releases what has always been the critical brake on physical production: no matter how cheap machines and tools become, you cannot manufacture new workers. In this piece, I will assume no additional new technologies and no recursive self-improvement to superintelligence. Yet, even with these conservative assumptions, I will show that the economy stands to be profoundly transformed.

Why VAISI Exists by

The Virginia AI Security Initiative (VAISI) is a student club at the University of Virginia. Our one line mission is: “Mitigate catastrophic risks from advanced AI.” I’m writing this post to explain what this means, why we think it’s important, and what we plan to do about it.

[…]

The problem is not purely a technical one. Philosophy is important in determining whether AI models are moral patients, and if not, then determining when they will (if ever) be. Politicians and policymakers need to set up institutions capable of regulating transformative AI, and will have to make critical, high-stakes, and rapid decisions as AI capabilities continue to progress.

What is the University’s role in the Commonwealth AI Institute? by Melody Yuan

“AI is too big and too pervasive for any one university, company or agency to tackle it alone. We wanted to find a way to bring higher education, industry and government together to cooperate around responsible AI,” Barnett wrote.

Subscribe now

The code is on GitHub. Starting a paper takes one command:

quarto use template stephenturner/quarto-arxiv-typst

I prefer Typst over LaTeX for speed and reproducibility. Typst compiles in a fraction of the time, and the error messages are usually more helpful, pointing at the actual problem instead of some long cascade of control sequences I can’t decipher.

R and Python chunks execute through knitr. kable and gt tables, figures from R and Python, numbered equations, and cross-references all render without extra work. You write the analysis and the manuscript in one file, render, and get a preprint-ready PDF.

YAML front matter handles the tedious parts for author metadata. This includes multiple affiliations per author with numeric superscripts, linked ORCID badges, and equal-contributor and corresponding author marks. An anonymous submission mode swaps in “Anonymous Author(s)” and turns on line numbering for double-blind review, and a hide-emails option keeps addresses suppressed. Citations use bracketed numerics through a bundled natbib CSL.

Install it with quarto add stephenturner/quarto-arxiv-typst, or copy the _extensions/arxiv directory into a project by hand.

More info at the README: github.com/stephenturner/quarto-arxiv-typst.

Subscribe now

I used Claude Fable to vibe code a little tool to take a structured text file and turn it into a QTI zip file you can upload into Canvas.1 All data stays on your computer. Everything happens in the browser. No AI involved.

Web tool: https://stephenturner.github.io/csv2canvas/

Source code (MIT): https://github.com/stephenturner/csv2canvas

You create a CSV with your questions, answers, and correct answer, and drop it into your browser. If the format is correct, it’ll preview what the quiz questions and answers look like. The page provides example CSV files to use.

Then you download the QTI zip file, head over to Canvas, and import the questions.

Everything happens in the browser, and your data never leaves your computer. The page provides example CSV files to use. But, if you have questions and an answer key in a word doc, the page also provides a button you can click that’ll copy a detailed prompt to your clipboard that’ll let you use your LLM of choice (local or frontier, doesn’t matter), provide your questions/answers, and it’ll create the spreadsheet for you to upload into the tool. YMMV on this one, but I’ve found it to be pretty good.

The fine folks who pay my Claude bills told me this cost about $25 in API credits. I ended up taking the code to GPT-5.6 Sol in Codex to do some UI polishing. So call it $30. Expensive for ~3 requests and responses. But dirt cheap considering how many hours of my (decidedly not cheap) time this will save.

CSV to Canvas app

CSV to Canvas Source Code

Subscribe now

A week or so ago, a new pre-print asked 165 LLMs to name a random number between 1 and 100, as well as some other questions that have single token outputs, and showed that it’s possible to predict with 60% accuracy which model family was used based on these set of tokens.

If the work is good why does it need a story? by

Regardless of how interesting and logically sound your work is, you will have to craft it into an engaging story if you want to reach a larger audience. The purpose of storytelling is not to manipulate. It is to overcome your audience’s inertia, most importantly boredom and disinterest.

LLM-enabled pandemics are fiction, for now. Don’t let them become fact. by

Let me be clear: we are not yet at the point where anyone can vibe-code a novel, stealthy high-risk Potential Pandemic Pathogen (PPP). The headline AI-assisted synthetic viruses are innocuous phages, and in fact quite similar by sequence to the parental virus. In the future, novel high-risk pathogens might be a prompt away. For now, and in my estimation for the near future, it’s not. Perhaps it’s never possible.

Giving the AI models a biology exam by

And this is the final line and the one most relevant to questions of biosecurity: Anthropic’s models may be doing lots of refusals, but if you want to get around them, you have lots of other options. Your accuracy may be reduced by some 10-20 percentage points, but you may pay as little as a fortieth of the cost to still get some pretty smart AI biologists.

Improving Fable 5’s biology safeguards by Anthropic

In our testing, this update reduced biology-related fallbacks by about 85% across our product surfaces.

[…]

It’s often difficult to tell apart beneficial and harmful uses of AI in biology. For example, in some cases researching a treatment for a disease requires scientists to produce the dangerous compounds that cause that disease in the first place. This is most obvious for live vaccines, which require scientists to grow the same pathogen they’re aiming to prevent.

Why AI-assisted bioweapons won’t kill us all by

Since the 1980s, there has been exactly one fatal bioterror attack.

[…]

AI can supply biology and chemistry knowledge, but not hands-on skill.

[…]

Reading every surgery textbook doesn’t make you a good surgeon. The biology laboratory works in a similar apprenticeship model. You learn by failing repeatedly under a mentor who has also failed repeatedly.

Bioweapons as the Optimist’s Exception by

She is right that bioweapons are hard today. She is wrong to infer that they will stay hard, or that this justifies being, as she puts it, “no longer worried” about bioweapons.

[…]

Unlike cybersecurity, biosecurity has a profound offense-defense asymmetry. An attacker does not have to go through clinical trials or pay for manufacturing. An attacker has time to design a weapon and can choose when to release, but the defender can mostly only start responding after the attack.

Benchmarking is not validation by

Benchmarking asks whether a tool performed well on a defined dataset at a point in time. Verification asks whether a tool continues to do what it claims to do, across versions, inputs, and edge cases. Clinical validation asks whether the result is fit for the medical use in front of it.

On Fermi Problems in Human Genetics by

These skills are now more important than ever, in this new era of vibe-coding. Let’s say Claude Code confidently gives you some answer. How would you know if its answer is even in the right ballpark?

The only reason you’ll ever need not to write with AI by Colin Carlson

Academic writing is a collaborative venture, and in my experience, very few researchers have a comprehensive conversation about GenAI use before they collaborate. [...] That means that all of us have the ability to end each other’s career.

LLMs reward expertise by Sean Goedecke

For many tasks, the human is the bottleneck, not the model, because the difficult part is in communicating to the model exactly what kind of solution the human wants. The information is “in the model” already, but it takes a very smart human to pull it out.

On Biological Data Generation 2/n: The Economics of Metrology by

The unit I want to reason about is therefore the cost per task-relevant, identifiable observation. An observation is identifiable when it preserves enough of the relevant state and context that the latent variable, transition, or mechanism can be inferred without a cascade of heroic assumptions.

“Code was never the hard part” is an insult to all programmers by Senko Rašić

If coding is easy, how come programmers were in high demand, and have demanded large salaries for years (even before ZIRP)? Why was there so much stress, overwork and burnout even before AI started churning out 5000-line PRs?

If coding is easy, why is software so damn buggy?

Whoever you are, don’t outsource your understanding, judgement, empathy and taste to AI. Don’t abdicate your responsibility. Don’t be a meat proxy.

Don’t be a meat proxy by Niklas Gruhn

This is not adding value. I can talk to Claude myself. It’s going to be faster and I get to control the context. I don’t need a meat proxy in between.

Why Is Everyone In Tech So Sad? by Aaron Horwath

What’s different about this moment is that the questions are not just economic but existential, the kind of questions that cause high-earning technical professionals to contemplate throwing it all away to start a goat farm in Washington or become a surf instructor in Costa Rica.

The Quest For Caffeine You Can Have At Night by

Since the beginning of time, mankind has yearned to drink a cup of coffee at 5 PM and go to sleep at 10. But now there’s an entire scientific subfield and several companies pandering to this absurd fantasy.

If there’s a war on cars in the USA then the cars have won, by Max Mautner

Drivers struck and killed 7,080 people walking in the United States in 2024, 72% more than in 2009, when pedestrian deaths were at their lowest.

I’m Begging You: Never Write With A.I., by Bret Stephens

What, uniquely, does writing do? It compels thought. It compels thinking in ways that silent contemplation or spoken language rarely can. It compels us to subject our thinking to the effort of articulation, the rigor of grammar, the tests of intelligibility and coherence, the inspection of others. In doing so, it also enjoins us to be clear, logical, accurate — and accountable.

Subscribe now

Even Elite Colleges Have Become Trade Schools (gift link), by Jessica Grose

Young people who don’t have the resources to pay out of pocket for their college degrees are going to be pushed into jobs that they might be ill suited for, rather than getting to explore their intellectual potential in fields that are not immediately remunerative.

My own thoughts on this one, starting with a deliberately provocative response:

Maybe elite colleges should be trade schools.

At least in part. The author argues that elite colleges already are trade schools and she treats this as a loss. Her worry is that students get pushed toward jobs they might not be suited for instead of being able to “explore their intellectual potential in fields that are not immediately remunerative.”

Some context on where I’m coming from might help.

For y’all who don’t know me well. I’m a first generation college student from the hollers of a rural county in southern Virginia. We had one stoplight. There are now more MAGA signs than intersections. Epithets like Redneck and Hillbilly were terms of endearment (unless you were an outsider, then them’s fightin’ words). My mother worked in payroll and billing at a local clinic. My father was a machinist and then an HVAC maintenance mechanic at a Westinghouse (later Hitachi) factory that built electrical transformers (a trade whose demand and pay have climbed hard in the few years since he retired). Some days I envy it actually. An honest day’s work with your hands. The job stays at the plant when you clock out. No inbox at 9pm. I love fixing stuff, from cars to bikes to all the weird stuff in the 100-year house I now call home.

Now we weren’t poor by any means. My parents poured every bit of savings they had into me, but it was nowhere close to being able to pay for my college tuition outright. I was in that awkward middle, not wealthy enough to pay for college outright, but not poor enough to qualify for need-based scholarship. I took out student loans to attend James Madison University (a state public school here in Virginia) and I was saddled with debt that I had to pay down for over a decade after finishing. College was a path to earn a better living than my parents, and their parents before them. And it had to be. I didn’t come from that Old Southern Money that people think about when they think of old Virginia.

Some back of the envelope arithmetic, using public information. The public state university where I work publishes an estimated total cost of attendance for 2026-2027 that runs from roughly $43,000 to $55,000 for in-state students, depending on entering year and school (scroll down and expand any of the Schools). Let’s make it easy and round it to $50,000. Four years comes to about $200,000 for a public state school in-state, and closer to $400,000 from out of state.1

Studying whatever interests you, finding your way, exploring your intellectual potential, etc., with no regard for what the labor market will pay… this is a luxury good at this price. An 18 year old kid has no way to price the bet they’ll have to take to finance this with borrowed money, and I’d consider it borderline unethical if I were to advise a student to take on $200-400k in debt for a credential with no plausible repayment path, unless they come from Family Money. Maybe it’d be different back in nineteen-diggity-something when you could buy a house for a nickel and gas was free. But alas, 2026 is different and college ain’t cheap.

None of this is an argument against the humanities and arts. In fact I’d highly encourage anyone headed off to university to spend as much time as you can outside your major. My most memorable and impactful classes I took as an undergraduate weren’t in my STEM major. One was an Eastern Religions course I took in the Department of Philosophy & Religion. Another was a Human Sexuality and Intimacy course in the Psychology department, taught by one of JMU’s most beloved professors, Uncle Bijan (a course that incidentally I took with my now wife). I had actually planned to go to JMU to double major in music, until I got bitten in the face by a large dog in the summer before my freshman year, tearing my lower lip in half, making trumpet a thing I’d never be able to play again.

I wrote in December about the false choice between meaning and accountability in higher education, and I still stand by it.

If you’re going to pay $200-400k for college, that college owes its students both serious intellectual formation and serious vocational preparation. Asking which of the two a given program actually delivers is just basic accountability, and it doesn’t make you a philistine. You’d prefer to get both, but if you don’t come out with employable skills2 enabling you to recoup those costs, you’d better come with a trust fund or an athletics scholarship behind you. If finding your way or self-discovery is the only thing that matters to you, there are many other options. Enlist in the military. Walk the Camino de Santiago. Do a few years in the Peace Corps. Hike the AT with the library of congress on your 200-gram e-reader. Coming of age and finding yourself doesn’t have to cost a small fortune.

The author of this essay ends by rooting for the Lehigh graduate she profiles, who described job hunting as riding a bicycle blindfolded. I’m rooting for him too. I’m also rooting for the kid from a backwoods county with one high school who will be the first in her family to hold a bachelor’s degree. She should get the art history seminar and the marketable skill. Both-and. What she can’t afford is a quarter million in student loans and a shrug about what comes after.

Subscribe now

Anthropic hasn’t published its algorithm. As I understand it: Before each token, a hash of the preceding token or two plus a secret key seeds a random number generator that splits the vocabulary into a green list and a red list. Green tokens get a constant added to their logits, biasing sampling toward green without forbidding red. Detection counts green tokens and runs a prop test (z). A widely shared explainer from GPTZero’s CTO was pretty informative to me.

Codon bias is the closest biological analog. Third position wobble lets you encode the same protein many ways, and a bias toward particular synonymous codons is invisible in the translated product but obvious in a chi-square test over a couple hundred codons. Except in AI text watermarking the preferred set gets redrawn at every position from the preceding context, so there is no fixed table of watermark words and no synonym swap that reverses it. So if you think you can quietly shake a watermark by removing the load-bearing em dashes and it’s not X it’s Y — the honest thing is I don’t think this is going to work.

Most of the reaction on Hacker News defends against a different mechanism than the one Anthropic described. Commenters assumed the mark has to be zero-width Unicode or nonstandard space characters, then proposed grep and git hooks to strip it. Neither touches a statistical watermark, because nothing anomalous appears in the character stream. Simon Willison raised a bigger concern that I share. When he asks Claude to refactor a specific function he knows exactly what he wants, and he doesn’t want a pattern baked into the result to satisfy a some voluntary EU commitment.

A detected mark means less than what I think most people will assume. Anthropic’s documentation says it indicates content was processed by Claude, and lists proofreading, translation, and summarizing as cases where the words came from a person. This probably hits some groups harder than others, namely non-native English speakers, or even writers with dyslexia who run their own drafts through an LLM before sending them. Paste your paragraph in for a grammar check and what comes back is marked, statistically indistinguishable from a paragraph the model wrote from nothing. You didn’t add the mark, you can’t inspect it, and stripping it means changing roughly a quarter of the tokens. Ian Miers, a co-author on the original green-list paper, posted that a watermark can’t separate a human writing with assistance from a machine writing alone, and that in 2026 “pure human text” feels like a relic.

The failure modes cut both ways. Content Claude touched at any point can return positive, content that returns negative can still have come from Claude, and fully human-written text can in principle be flagged. Anthropic’s documentation acknowledges the first two.

After aborting AI Dry July I landed on a rule: prose I’ll have to defend is mine to write, and anything with an “I” in it gets no LLM assistance, ever. Watermarking doesn’t change what I do. A detector firing on a document says nothing about which sentences came from where, and the institutions most likely to run one, journals and funders among them, are probably the least equipped to read the result. Anthropic is explicit that absence proves nothing and presence isn’t conclusive, but I’m afraid this caveat will get memory-holed by next week.

Subscribe now

Please see the job ad, description copied below. More may be on the way. The position is in-person, here in lovely Charlottesville, Virginia.1

Apply now!

Contact me for questions

Apply now!

I also have another job ad open, but will be closing this one soon: Postdoctoral Research Associate, Machine Unlearning and Model Editing for AI Biosecurity.

Share

Subscribe now

Granola solves this problem (and a bunch of other problems I don’t have) for a monthly subscription fee. I built Trailmix to help with this for free, with everything staying private on my machine, no third party or monthly fee involved.1

Get Trailmix

Trailmix is a MacOS menu bar app that records system audio, your microphone, and optionally a window or a whole display, then transcribes it all on-device to give you a text file transcript, and mixes both audio channels into a video mp4 you can watch later. Nothing is uploaded anywhere, there’s no account and no API key, no monthly fee, no cloud, no AI (yet). There’s no app window either, just click the Trailmix icon, pick your sources, and start recording.2

Features

It keeps two audio tracks separate. mic.caf is your microphone, system.caf is everything the Mac played. Each gets transcribed on its own and merged by timestamp, so transcript lines are tagged Mic or System without any speaker-identification / diarization model involved. You get a combined mixed.m4a audio recording, plus a mixed.mp4 with the video and audio if you captured a screen.

The Stop after feature is for the webinar case. Arm a recording, walk away, and it closes itself out at the time you picked (a preset, or e.g. type 1h10m). If the meeting runs long there’s a menu item to Add 15 minutes. Recordings stop after 8 hours regardless, so one you forgot about can’t fill the disk overnight.

Transcription is done on device with Parakeet TDT 0.6B v2 through FluidAudio’s Core ML port, around 20 seconds per hour of audio on Apple Silicon. The 440 MB of weights download once. There’s also an on_stop hook that runs a shell command on the session folder when the transcript is done (for example, you could wire this up to move files around or hand the transcript to an AI for a summary).

Roadmap

On the near-term roadmap: summaries generated locally too, next to the transcript, with key points, decisions, and action items. I’m not going to re-read 8,000 words from an hour long meeting, and shipping off all those words to a frontier model API undoes the reason I’m doing everything local on my own machine in the first place. Next thing I’m bringing to this is the ability to (optionally) generate AI summaries locally with llama.cpp using a small open-weight model running on your machine.

Other miscellaneous features I’d like to implement: noticing and warning about a silent audio track while it’s still recording; adding a search feature to read through transcript.json files across the recordings folder; add a recent sessions selector in the menu; optional notifications to offer to record when a different app requests the microphone; decoupling the computer audio capture from the video target; configurable microphone selection and preflight checks; and transcription model selection beyond Parakeet to enable non-English transcriptions.

Distribution

Right now if you want to try it out you’ll have to build it from source. It’s just a git clone followed by a make install, and the README that walks through all of it.

Getting it into the App Store, or even just creating the .app inside a .dmg you can install yourself without MacOS Gatekeeper’s scary unidentified developer warnings — these require an Apple developer account at $99/year.3 So, a question: would you pay $10 for Trailmix? One time, for life, no subscription. If I can find a few people who would, I’ll pay Apple’s fees and do the signing and review work, and it becomes something you can install and forget about. Let me know in the comments or just email me.

https://github.com/stephenturner/trailmix

Subscribe now

The markdown on the left hand side is editable. You get a preview of the rendered markdown on the right hand side. You can copy or download the markdown, as well as copy the rich test or download a rendered HTML file.

Subscribe now

If you paste in a link it’ll grab that PDF and convert to markdown. The URL you get is shareable. E.g., you can paste in the URL to the PDF of the recent BioTIER paper at https://arxiv.org/pdf/2607.14479, and you’ll get back the rendered output with the shareable URL in your address bar: https://stephenturner.github.io/pdfdown/#url=https%3A%2F%2Farxiv.org%2Fpdf%2F2607.14479

Web app: https://stephenturner.github.io/pdfdown/

Source code: https://github.com/stephenturner/pdfdown

Subscribe now

Subscribe now

What I’m reading

AI company Anthropic announces it will begin developing drugs of its own, by Brittany Trang

Eric Kauderer-Abrams, the company’s head of life sciences, said Anthropic has been asking itself what it should be doing besides training models and building products. During an event here to launch the company’s newest application, Claude Science, he said Anthropic had come up with one answer.

“We’ve decided to start running some drug programs ourselves, and we’ve chosen to do this by running drug programs in the pre-clinical stage … and choosing indications for neglected disease,” he said.

Why AI will not speed up science (yet) by

The actual bottlenecks in biomedical science are physical, institutional, and human. Getting a grant funded takes 12 to 18 months from submission to award, and that assumes it is funded at all; most proposals require years of revision and resubmission. Recruiting human participants for a clinical study takes years, and enrolling a cohort with rarer characteristics can take longer still. The physical constants of biology are indifferent to how quickly a model can read: cell lines must be cultured, samples collected and shipped, sequencing runs queued and completed, and mouse colonies bred to the correct genotype over many months, with some disease models yielding a first meaningful readout only after the animals have aged. Experiments fail and have to be repeated, protocols need optimization, reagents and specialized instruments have lead times, and shared core facilities have waiting lists. Layered on top of all this is an institutional apparatus that runs on its own clock: IRB approval for human subjects and IACUC approval for animal work, biosafety approvals for pathogens, data use agreements and material transfer agreements with legal back-and-forth between universities, and controlled-access approvals for protected genomic data all take many months.

Pacing the Frontier, signed by >1,000 employees at Frontier AI companies

We request that the U.S. government support an international effort to develop the technical and governance tools needed to deliberately pace the frontier of automated AI development.

NIH Is Replacing Letters of Support with Letters of Collaboration to Reduce Administrative Burden (NIH Notice)

Letters of collaboration should be limited to stating the intent to collaborate or provide material support as described in the application and should not contain endorsements or additional information about or evaluations of the proposed project. An individual letter of collaboration must not exceed 100 words. NIH encourages applicants to use the following single-sentence format for letters of collaboration.

“If the application submitted by [Organization Name and with PD/PI Name] as Program Director/Principal Investigator entitled [Grant Application Title] is selected for funding by NIH, it is my intent to collaborate and/or commit resources as detailed in the application.”

‘OnlyMarms’: Marmots Are on OnlyFans to Raise Money for Research, by Emmet Linder at The New York Times

“A team of scientists uploaded videos of the squirrel-like rodents in the wild on the adult-content site OnlyFans to raise money amid government budget cuts. […] But Dr. Blumstein hit a snag when he had to upload a photo of his driver’s license to create the OnlyFans account. The platform notified him that the face in the license photo did not match the ones nibbling on leaves or frolicking across a dirt patch in videos and photos he’d uploaded to the site.”

Generative AI Availability, Grades, and Student Satisfaction at a Large University, by Dumlao et al (arXiv 2607.21534)

[when adjusting for the effects of the COVID-19 pandemic] We find no significant differential effect of GenAI availability on grades overall or among previously lower-performing students. Effects on self-reported understanding are likewise insignificant; […] Our findings temper concerns that GenAI inflates grades and reduces students' satisfaction.

Buying AI Is Easy. Becoming a Different Company Is the Hard Part, by

The AI conversation in biotechnology is currently dominated by models, vendors, partnerships, pilots, and announcements. Companies are buying licenses. Teams are experimenting with tools. Executives are being told that everything is about to change. And some of it will. But buying AI is easy. Becoming a different company is the hard part. A different company is not a different tech stack. It is a different culture: how people work, what they are willing to let go of, and whether the organization can learn a new way of operating and make it hold.

Behind the Curtain: AI titans’ biggest private fear, by Jim VandeHei and Mike Allen

We ask almost every AI architect and leader the same thing in private: What AI risk worries you most? Almost all of them fire back the same response: a killer pathogen, spreading too silently, widely and quickly to stop.

Artificial Intelligence Among University Professors, by Jean Fan

AI recommendation letters the same generic platitudes and observational summaries that could be simply read off a candidate’s CV, without any seeming awareness of the fact that: If I had wanted an AI’s opinion, I too am capable of prompting ChatGPT.

On the HuggingFace “Incident” and the biosecurity angle, by

the thing that distinguishes attacker from defender is context, and the AI models currently have no reliable way to see context […] Plenty of biologists get refused when they try to use Claude Fable to help with their research. A virologist doing outbreak forensics and the person trying to enhance a pathogen ask overlapping questions, and a content filter can’t separate them. If we build a biodefense ecosystem that depends on frontier APIs, and those APIs refuse to help during the emergency — the one time you actually needed them, at speed, on real pathogen data — then we will have built a defense that switches off precisely when attacked.

Who should be responsible for OpenAI’s hack of Hugging Face?

The same escape could instead have produced harms beyond anything the developer could pay: imagine OpenAI’s models had targeted critical infrastructure rather than a benchmark database. At that point, the threat of liability stops deterring.

Surfacing Benchmark-Maxxing in Kimi-K3 by at

results for Kimi-K3 […] across our short-horizon therapeutics and -omics benchmarks […] we observed a fixation on optimizing for the eval’s grader, with the model spending large amounts of time reasoning about a reference/golden solution and burning tokens to infer the evaluation writer’s intent. […] It is not clear other models do not display evaluation awareness or benchmark maximizing tendencies, it’s just that Kimi-K3 is much more explicit about it. We are fairly confident that other models are eval aware, but are less explicit about its exploits than Kimi-K3.

Human brain changes after first psilocybin use (Lyons et al, Nature Comm. 2026)

Increases in cognitive flexibility, psychological insight, and well-being are seen at one-month. […] Increased cortical signal entropy (EEG) at 1- and 2-hours post-dosing predicts improved psychological well-being at one-month. […] The present work sheds light on human brain changes under and after first-time high-dose psilocybin. The high-dose session was each person’s first-ever psychedelic experience. All—except one participant—rated their high-dose experience with psilocybin as the single most unusual conscious state of their entire lives. The one person who did not, ranked it within their top-five most unusual conscious experiences.

Who’s Afraid of Chinese Models? by Ben Thompson

This is a point that bears repeating: because U.S. open weight model makers must follow the frontier labs’ terms of service, they (1) are worse than Chinese alternatives and (2) end up distilling the distillation, just with a detour through Chinese labs. Wouldn’t it be better if western open weight model makers could go to the source?

When it comes to uncertainty, AI research is lagging behind by

LLMs are a promising tool for scaling data interpretation, but like many measurement tools before them, they are often also noisy, biased, and non-reproducible on their own. This means it’s crucial to characterise when and why they fail.

The AI Future Is for Everyone, by Mark Zuckerberg

Healthy societies also recognize that safety requires checks and balances. If only a handful of institutions have superintelligence, they will inevitably exercise a controlling influence over economics, science and politics. Even with the best intentions, that concentration would limit people’s ability to choose their own future.

An opinionated guide to which AI to use to do stuff, by

So my practical advice remains pretty similar: pick Claude or ChatGPT, pay the $20, and give an agent a real task from your real life. Then look carefully at what comes back, and, rather than just accepting or rejecting the results, ask for changes, just as you would ask a real person. See if you can accomplish your goals, even if you failed at first. You will learn more about what AI means for you from that one experiment than from any guide, including this one.

The Assault On Science Funding Continues, by Derek Lowe at Science

The Trump administration hates academic science funding, full stop. They hate where that money goes, and they hate who it goes to. They want to keep all that money for themselves, to hand out to favored cronies who can help them get elected and to steer yet more money and more power back into their hands.

University science in the US needs a coherent plan (H. Holden Thorpe, Science)

If universities oppose the actions of the administration, they need to say so quickly so that their constituents aren’t left wondering whether there’s a plan. The alternative is a community of university scientists adrift and grasping for a lifeline.

Open Weights and American AI Leadership, by a lot of people.1

In fact, openness may be one of the most important paths to AI safety and security. Relying solely on closed models is not inherently safe: they can be breached, misused, or fail in ways that outsiders cannot detect. And concentrating advanced AI capabilities behind a small number of closed models compounds that risk. It results in a small number of single points of failure, weakens competition, and leaves critical technology in the hands of a few providers. Open weight models, on the other hand, allow a broad community of researchers and developers to examine their behavior, identify vulnerabilities, develop safeguards, and improve them over time. Just as open-source software demonstrated that transparency can be more secure than obscurity, AI safety may depend on giving more people the ability to test and strengthen the models on which society relies. It allows for rigorous benchmarking and evaluation, red teaming, and protections tied to real and demonstrated harms rather than assuming that closed systems are safer by default.

The Rise of the “Scenario,” a Trendy Way of Forecasting our Dystopian Future, by Gideon Lewis-Kraus

The key characteristic of history, in this more refined account, is that it has a long and sterling track record of continuity. Certain things may change incrementally—electricity, antibiotics, democracy—but it is the habit of the future to more or less resemble the past. Given the propensity of today to look a lot like yesterday, the odds that tomorrow will look like today are overwhelming. As we have learned from intermittent financial crises, “this time it will be different” is almost always a dumb bet. […] The premise that the future will resemble the past because it has always resembled the past will continue to be true until the first time it isn’t.

Trump’s Plan for Science: More Money for A.I., Less for Universities, by Christopher Flavelle and Sheryl Gay Stolberg at NYT.

the administration would prioritize artificial intelligence, robotics and nuclear energy while reducing the government’s focus on what he called “life sciences.”

[…]

His call for the administration to de-prioritize “life sciences,” which encompass molecular biology, genetics and other fields in biomedicine, as well as clinical trials aimed at testing new therapies

[…]

It says federal agencies “should prioritize foundational research in the biological sciences over the life sciences.”

Who cleans up after the vibe-coding party? by Sam Learner

“Our findings frame AI slop as a tragedy of the commons,” the researchers concluded, “where individual productivity gains externalise costs on to reviewers, maintainers and the broader community.”

Wooly Mammoth startup Colossal seeking at least $20B valuation by Lucinda Shen

The lead investor could not be learned, but the round is expected to value the company between $20 billion and $30 billion.

Measuring Biosecurity Safeguard Effectiveness with BioTIER by Eleanor Marshall

The strongest guardrails remain concentrated in a few highly-capable closed models. Importantly, the capabilities of closed-weight models are only ahead of those of open-weights by around 4 months. This means that even the most robust closed-model safeguards offer little practical security if highly capable and permissive open-weight alternatives are freely available.

CRISPR gets a power boost from AI-designed ‘molecular scissors’ by Amanda Heidt

“This particular work illustrates the future of AI and biology, which is that we need AI tools to help us go faster, but we still need people who really understand molecular mechanisms if we want to make maximal use of these kinds of models,” [Jennifer] Doudna says.

Mind the alignment gap: a spatial transcriptomics benchmark for scientific coding agents by Yiqun Chen et al.

In this work, we constructed a semi-automated benchmark derived from an existing spatial genomics alignment paper and validated its use for evaluating general coding agents on biomedical tasks. Our key finding is that it is much easier to prompt agents to use specialized packages than to use them effectively: more package-oriented prompting increased package probing and code submissions, but agents were often nudged toward poorly tuned variants that scored lower than their own geometric heuristics.

What will be left for us to work on? by

Computers have often been called bicycles for the mind. I think AI can be more than that. I think AI can be a crane for the mind, if you will indulge my metaphor, in the sense that it can amplify our potential to previously unimaginable heights. Getting there can seem daunting. It has an incredible learning curve. I feel like I’m on a treadmill all the time, but I’m very excited about it. I think it’s a fun challenge, and I think it’s worth fighting this fight. Compared to five years ago, in a way, I feel ... maybe superintelligent is not the right word, but I feel like I have superpowers, given the extent to which AI allows me to take on new ambitious things that were not possible before, and push myself harder than was possible before.

The twilight of the chatbots by

The instability is what happens when institutions that move at the speed of people (or worse, committees) try to track a capability curve that is very much not human in nature. And as long as we are on some sort of exponential, and for as long as it lasts, the gap only widens.

From Confused to Contributing: My First nf-core PR Journey by Jimmy Lail

When I started this PR, I thought nf-core was a repository where you went to download pipelines. What I learned is that it’s an ecosystem and a community, and a genuinely welcoming one. Every time I got stuck, someone in the Slack community walked me through it. That’s why I’m now a Nextflow Ambassador, and why I’m writing this post. You don’t need to submit a PR to get started. You don’t need to understand everything before you try. Run an existing pipeline. Go through training. See if it clicks. The community is there when you need it, and the framework is better than it looks from the outside.

How useful are zero-shot predictions of mutational effects? by

In summary, zero-shot predictions capture protein viability, but they are rarely useful to identify function-enhancing mutations. […] Zero-shot predictions can be useful, in particular to screen out strongly deleterious mutations, but they will rarely point towards increased function in targeted protein-engineering campaigns.

Ancient DNA solves Medici murder mystery by Taylor Mitchell Brown

working on ancient DNA truly feels like looking directly through the keyhole of history.

Whoops! Most arXiv papers contain information never meant to be shared by David Brzostowicki

It found that 88% of submissions that contained LaTeX source files included some form of hidden information, from arguments between co-authors and to-do lists acknowledging weaknesses in the text, to passwords, GPS coordinates that can reveal a researcher’s home address, and application programming interface (API) keys

Mass-produced science is coming. What happens to scientists? by Kenneth Harris

Biologists and other experimental scientists might experience a near-term boom like the weavers. The efficiency of AI science could lead to a surge in science investment, for example as treatments for previously untreatable diseases come into reach. This could lead not only to benefits for patients but to a rise in demand for biological labor, at least until these roles too become automated.

Publishing Without Journals: An Open, Forkable Archive with Attributed Review by Matthew Lorig

The journal is a seventeenth-century technology asked to do four modern jobs at once: disseminate results, certify their quality, allocate scholarly attention, and confer career credit. It does none of them well.

A data bottleneck could slow the superintelligence race by

AI R&D consists of a myriad of skills that each take real-world experience to learn — writing clean code, designing experiments, research “taste,” coordinating across dozens of parallel research agendas, optimizing data-generation pipelines, running compute clusters the size of small cities, producing the chips that run them, and so on. Even leaps in sample efficiency can’t fix this, because the data simply doesn’t exist yet.

The AI Superforecasters Are Here by

AI forecasters are the same kind of advance as going from a world where writing required hiring a scribe and baking a clay tablet, to a world where writing only requires hitting the “send tweet” button.

Synthetic biology and nature conservation by IUCN

synthetic biology could open new opportunities for nature conservation. For instance, it may offer solutions to currently unsolvable threats to biodiversity, such as those caused by invasive alien species and diseases.

GCBR Organization Updates, July 2026 by and

• What the UN, WHO, and UK government are saying about mirror life governance;

• 88 countries convene in Kuala Lumpur for GHS2026 amid a growing Bundibugyo virus outbreak;

• Blueprint Biosecurity launches its first RFP on pathogen-agnostic biothreat detection, plus new jobs and fellowship deadlines across the field;

How I think about catastrophic biological risk (part I): risk breakdown by type of response by Andrew Snyder-Beattie

Overall my take is that the vast majority of biological risk comes from pathogens that directly infect and kill humans. Specifically I’d say that >95% of the risk is coming from attacks that target humans (both direct and indirect risk), with less than 5% of the risk being attacks that target agriculture, and less than 1% of attacks that target the environment.

How I think about catastrophic biological risk (part II): risk breakdown by type of prevention by Andrew Snyder-Beattie

In general, the information hazard tradeoff typically isn’t worth it because there are not severe risks associated with the well-intentioned research (the one big exception being mirror bacteria).

10 big projects for reducing bio x-risk by Chris Bakerlee

Yet remarkably few people are working full-time on this problem. By my count, there are ~160 people on the planet whose full-time job is reducing bio x-risk. This entire group could fit on a single short-haul flight.

Making CAISI the AI agency we need, by

According to two sources familiar with CAISI’s current role, it continues to proceed with frontier model testing and research, as it has always done since it was created under the Biden administration. Though small, it is widely thought to have some of the strongest technical talent in Washington, including former OpenAI model alignment lead Paul Christiano, who is head of AI safety. The problem is, nobody in the federal government or at the frontier AI firms is legally obliged to listen to them. That means that even though the United States produces the world’s best AI models, its regime for vetting those systems has little influence over real outcomes. […] Compare that to what’s happening across the pond. The UK’s AISI is its government’s primary office for overseeing frontier AI, and it has almost six times the funding of CAISI, with more than three times its staff. Technically CAISI and AISI have a memorandum of understanding to share research and model testing, but the UK institute appears to have been more adept at doing that work, routinely finding vulnerabilities other government agencies overlook. OpenAI’s model card for 5.6 and Anthropic’s system card for Mythos and Fable both list multiple vulnerabilities found by AISI, and none from CAISI.

Subscribe now

What I’m watching

Subscribe now

What I’m writing

My five most-read newsletters this month.

Subscribe now

Hanna Palya and Cassidy Nelson propose ADAPT (Advanced Detection of AI-enabled Pathogenic Threats) as a five-year successor to the IARPA FunGCAT program that ran from 2017 to 2022.

The case for ADEPT is that the field has no operationalizable definition of what makes a sequence concerning,1 which leaves providers interpreting vague guidance and regulators unable to enforce some floor for “concern.” And when current tools detect threats by sequence similarity rather than function, a redesigned virulence factor might slip past these filters, being read as benign.

As proposed, Phase I would replace static watchlists with a weighted, hierarchical ontology that scores sequences across attributes like molecular function, necessity for pathogenicity, host range, and weaponization potential. A housekeeping gene shared with a pathogen scores low, but a virulence factor essential for host harm would score high, even after its sequence has been redesigned past recognition. Phase II would build the tools as cascades, where cheap filters screen everything and expensive protein language model analysis handles only what makes it through to this point.

The hard part (as usual) is annotation. Pathogenic functions might sit on peripheral nodes of a Gene Ontology graph, and CAFA benchmarks show function predictors aren’t that great there. Palya and Nelson lean on protein language model embeddings, few-shot classifiers, and graph neural networks over GO to populate the ontology faster than manual curation allows for. They cite few-shot work that generalized to new toxin classes from tens of examples, which would let screening begin before large training sets exist. Budget estimate: $65 to $80 million, i.e., between an ARIA and a DARPA program.

Full disclosure: I worked on the IARPA FunGCAT program for a few years with SigSci and friends. Throat-clearing out of the way: if you’re listening out there, IARPA (and I’m sure you are), I’d love to see a FunGCAT follow-on program like the ADEPT proposal here get some attention (read: $$$) from the likes of IARPA, DARPA, or even a potential DOE Genesis Mission-like program.

Subscribe now

I got an email today from ARPA-H with the subject, “Next on your summer reading list: the ARPA-H 101 series 🏖️”. If you’re ARPA-curious (ARPA-H specifically, but the advice is applicable to DARPA and IARPA as well), read through some of the links below to get a sense of how things work here.

You can sign up for the ARPA-H Vitals emails here.

Doing business with the agency: Working with ARPA-H differs from traditional grant-funding federal agencies. The expectations for continued performance are high, the timelines are aggressive, and receiving continued funding is dependent on milestone achievement. See how you can rise to the challenge.

Preparing for Proposers’ Day: These big events set the stage at the launch of most ARPA-H programs. Ready to make the most of networking and teaming opportunities? Study up and arrive prepared.

Assembling a research team: ARPA-H programs ask performers to offer solutions that are not evolutionary but revolutionary. That often means assembling teams of talented groups and individuals. Learn how we help connect them.

Transformative research needs Other Transactions (OTs): An OT is a type of federal contract that provides a more flexible way to fund innovative research and development. ARPA-H uses OTs to compress research timelines and speed real health impact. Explore the basics of this nimble funding mechanism.

The lifecycle of an ARPA-H program: The ARPA model, championed originally by the Defense Advanced Research Projects Agency (DARPA), centers ambitious ideas, accelerated research timelines, and breakthrough advances. Not familiar with this transformative model and how programs move through it? Read on.

Understanding proposal evaluation criteria: ARPA-H was created to close the gap between scientific findings and the urgent need for real patient outcomes. That means the bar looks different from a traditional grant application for other federal research funding agencies. Here’s what to expect.

Subscribe now

The last of the five features was fragility. In theory, a container is supposed to make that question moot. I.e., you package the dependencies, freeze them at build time, and the tool runs the same way years later. In theory.

Our new preprint, led by , tests whether that promise holds.

Nagraj VP, Turner SD, Magee N. Survey and Evaluation of Applied Containerization Practices in Bioinformatics. bioRxiv, 2026.07.17.739198. https://doi.org/10.64898/2026.07.17.739198.

We pulled 252 containerized bioinformatics tools from software papers published between 2021 and 2025, annotated how each one was built and maintained, then tried to rebuild and pull every image on a purpose-built Kubernetes platform, socr8s. Which, by the way, we made available at github.com/vpnagraj/socr8s (MIT).

Of the 195 tools with a build specification, only 50 rebuilt successfully. That’s only 26%, even after we retried failures to correct for build-context mismatches and transient network drops! The single strongest predictor of a successful build was not multi-stage builds, pinned parent images, or multi-architecture support. It was recent commit activity. Tools with a commit in the preceding two years were about twice as likely to build (OR 2.26, 95% CI 1.16-4.42).

Pre-built images held up better. Of 173 images documented on a public registry, all but 3 pulled cleanly. A registry copy is a real safeguard, but it preserves a tool exactly as it was pushed, so any security patch or compatibility fix means rebuilding from a specification file that may no longer work.

We used AI once, to triage build-failure logs with a Claude skill, and we manually re-audited its classifications and reassigned several. The failures were mostly bitrot: broken dependency chains, vanished parent images, repository restructuring that left COPY targets missing. Of the 10 failures the skill flagged as trivial to fix, we resolved 5 on the first try. The rest surfaced a second problem once we cleared the first.

Data, analysis code, and the preprint’s Quarto source files are here on GitHub.

⁂

TL;DR: Containerizing a tool is necessary but not sufficient. Without an update or a new commit now and then, the image slides toward the rot and irreproducibility that the container was meant to prevent. I.e., the fragility signal from SNIFF.

Subscribe now