Open tabs (June 12, 2026)

TBR in AIxBio, AIxEdu, AIxLabor, AIxWriting, and other essays & papers

I’ve had a busy week, taking the day off today, and I haven’t had a chance to do much reading. I’ve been spending a ton of time lately developing a new course I’ll be teaching this fall, and preparing a workshop on AI-powered literature review and synthesis I’ll be teaching next week (if you’re at UVA, register and attend for the in-person event if you can — it’ll be much more engaging than Zooming in, trust me).

Here are my open browser tabs I have open that I hope to catch up on soon.

Blogs/newsletters/etc

1. Dario Amodei — Policy on the AI Exponential

2. Paving the way for agents in biology \ Anthropic¹

3. NIH RFI on limiting the number of grants per PI

4. Claude Fable 5 and Claude Mythos 5 \ Anthropic²

5. Eight Predictions for the Future of Higher Education

6. Matt Lubin: Five Things: June 7, 2026

7. Is AI More Expensive Than the Employees It's Replacing?

8. The Anti-Scaling Law in Biology, and Why AI Could Make Crowding Worse Before Making Drug Development Better

9. Sasha Gusev: Thoughts on AI in academia

10. How LLMs Actually Work | 0xkato

11. A cell is not a spreadsheet- why “Virtual Cells” are still mostly hype

12. Anthropic: When AI builds itself

13. Can A.I. Produce Writing That We Actually Want to Read?

14. Is a compute crunch coming?

15. Built to benefit everyone: our plan | OpenAI

16. Bots are scraping open data — how should researchers respond?

17. Why Are Cells Small? - Niko McCarty

18. How to build a cancer vaccine, and whether they will work this time

Papers

1. The total eclipse of bioinformatics: From disruption to convention, and a gentle warning

2. Dual-use artificial intelligence and biology: upstream risk-benefit reviews

3. Molecular de-extinction looks to the past to find the molecules of the future

4. AutoScientists: Self-Organizing Agent Teams for Long-Running Scientific Experimentation

5. Pleiotropic shared heritability quantifies the shared genetic variance of common diseases

6. Ten simple rules for teaching data science

7. Depth normalization for single-cell genomics count data and Lior’s explainer

1. I just read this one right before posting. The post describes the difficulty agents have at retrieving biological data. Which isn’t limited to agents! It’s difficult for a human to navigate the disparate databases and web interfaces and NCBI Virus search incantations to get the thing you’re looking for. If this problem were solved for agents, it’d make life easier for us humans as well. A conclusion from the post: “We want models to be creative when they generate hypotheses, design experiments, or reason about mechanisms. But the layer underneath that creativity—gene identifiers, schemas, retrieval logic, coordinate systems, metadata conventions, and data access paths—has to be boringly reliable (or in other words, deterministic)”.

2. I haven’t had a chance to do anything with Fable yet, mostly because I work in AIxBio, and Bio is off limits. And because I’m a biologist, Fable refuses to talk to me (“Who am I?” leads to safety flags and demotion of the rest of the conversation to Opus). Precautionary principal is probably the right move here given the benchmarks, and I think managed access will likely be the way these models are released from here out.