$1M in 1995: - Buy forever home - Raise 3 kids - Stay at home spouse - Retire early $1M in 2026:

Thinking of AI as a productivity booster for prior workflows is the wrong framing. Like all of the previous waves of computerization/softwarization, AI is a tool that lets you do new things in new ways.

Les données disponibles sur data.gouv.fr sont désormais interrogeables via un serveur MCP dédié en experimentation, vos retours sont bienvenus ! 💻 Le code est ouvert et accessible sur GitHub : github.com/datagouv/datagouv… Pour en savoir plus : data.gouv.fr/posts/experimen…

Faculty and researchers from the UW Department of Ob-Gyn Division of Reproductive Sciences shared their work at the @SRIWomensHealth 2026 annual meeting, including an award-winning presentation! Read more about their accomplishments: obgyn.wisc.edu/ob-gyn-depart…

In AI-guided discovery, models often turn huge candidate pools into shortlists for costly validation. We ask: can we put an error budget on AI-generated shortlists before running the experiment? For example: • Can we keep failed hits below 10%? • How many candidates should we test to get enough true positives? • How far down the list can we go before expecting too many false positives? • If we already have a fixed top-K list, how many are likely wrong? 📢 Excited to share TxConformal, a framework to turn AI scores into shortlists with controlled/estimated false positives, even in tasks where new candidates differ from past experimental data. This is joint work with amazing @KexinHuang5 @jure @EmmanuelCandes , in collaboration with Genentech @nate_diamant @gabo_scalia. We test it across proteins, genetic perturbations, regulatory DNA, clinical trials, ADMET, and antibacterial virtual screening. In a prospective A. baumannii screen at Genentech, TxConformal estimated 80.3 false positives before wet-lab validation; the experiment found 91, within the 90% CI. Preprint: biorxiv.org/content/10.64898… Code: github.com/ying531/TxConform… 🧵[1/n] 👇

Judging by my tl there is a growing gap in understanding of AI capability. The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere last year and allowed it to inform their views on AI a little too much. This is a group of reactions laughing at various quirks of the models, hallucinations, etc. Yes I also saw the viral videos of OpenAI's Advanced Voice mode fumbling simple queries like "should I drive or walk to the carwash". The thing is that these free and old/deprecated models don't reflect the capability in the latest round of state of the art agentic models of this year, especially OpenAI Codex and Claude Code. But that brings me to the second issue. Even if people paid $200/month to use the state of the art models, a lot of the capabilities are relatively "peaky" in highly technical areas. Typical queries around search, writing, advice, etc. are *not* the domain that has made the most noticeable and dramatic strides in capability. Partly, this is due to the technical details of reinforcement learning and its use of verifiable rewards. But partly, it's also because these use cases are not sufficiently prioritized by the companies in their hillclimbing because they don't lead to as much $$$ value. The goldmines are elsewhere, and the focus comes along. So that brings me to the second group of people, who *both* 1) pay for and use the state of the art frontier agentic models (OpenAI Codex / Claude Code) and 2) do so professionally in technical domains like programming, math and research. This group of people is subject to the highest amount of "AI Psychosis" because the recent improvements in these domains as of this year have been nothing short of staggering. When you hand a computer terminal to one of these models, you can now watch them melt programming problems that you'd normally expect to take days/weeks of work. It's this second group of people that assigns a much greater gravity to the capabilities, their slope, and various cyber-related repercussions. TLDR the people in these two groups are speaking past each other. It really is simultaneously the case that OpenAI's free and I think slightly orphaned (?) "Advanced Voice Mode" will fumble the dumbest questions in your Instagram's reels and *at the same time*, OpenAI's highest-tier and paid Codex model will go off for 1 hour to coherently restructure an entire code base, or find and exploit vulnerabilities in computer systems. This part really works and has made dramatic strides because 2 properties: 1) these domains offer explicit reward functions that are verifiable meaning they are easily amenable to reinforcement learning training (e.g. unit tests passed yes or no, in contrast to writing, which is much harder to explicitly judge), but also 2) they are a lot more valuable in b2b settings, meaning that the biggest fraction of the team is focused on improving them. So here we are.