If you think about how statistics works it’s extremely obvious why a model built on purely statistical patterns would “hallucinate”. Explanation in next tweet.

Demis Hassabis is calling on philosophers to step in. He says technologists shouldn't decide AGI's future alone. “We need a new Kant or Wittgenstein to help map out where society should go next.” Politics, ethics, and theology will be essential to navigating a post-AGI world.

Looking forward to speak at Yale University next friday.

Register today for the Virtual Cell Challenge and use AI to solve one of biology’s most complex problems. Announced in Cell, the competition is hosted by Arc Institute and sponsored by NVIDIA, 10x Genomics, and Ultima Genomics.

PDF preview available here (no paywall): nature.com/articles/s4146…

Grok 4 is at the point where it essentially never gets math/physics exam questions wrong, unless they are skillfully adversarial. It can identify errors or ambiguities in questions, then fix the error in the question or answer each variant of an ambiguous question.

Our new Cell review on the generative A.I. era in medicine Pranav Rajpurkar cell.com/cell/fulltext/…

1/11 Excited to share our @Naturestudy led by Leon Ooi Csaba Orban Shaoshi Zhang doi.org/10.1038/s41586… It is well-known that AI performance scales with logarithm of sample size (Kaplan, McCandlish 2020), but in many domains, sample size can be # participants or # measurements...

Patterns, Predictions, and Actions:

“This is a gamechanger for the field.” A study co-authored by professor of neurology Nico Dosenbach with NUS shows how to optimize brain scans for designing smarter and more cost-effective studies of neurological and psychiatric conditions.

How do rare genetic mutations that affect multiple genes simultaneously shape brain development and behavioral outcomes in adolescents beyond just cognitive deficits?Nature Biomedical Engineering Mila - Institut québécois d'IA "A pattern-learning algorithm associates copy number variations with brain structure and

What are "omics"? A good schematic cell.com/neuron/fulltex… could add epigenomics, immunomics, microbiomics, glycomics, exposomics, etc

Great workshop - thanks again for inviting me to Banff.

We're launching an "AI psychiatry" team as part of interpretability efforts at Anthropic!  We'll be researching phenomena like model personas, motivations, and situational awareness, and how they lead to spooky/unhinged behaviors. We're hiring - join us! job-boards.greenhouse.io/anthropic/jobs…

📢 job opening: software developer at Mila Quebec AI Institute & McGill As of now, my team is looking to hire a new technical staff member with abackground in a STEM area, such as computer science, engineering or physics, to support core lab activities around data analysis,

Why #language #models hallucinate? ->We have experimental data on that. We precisely locate where #hallucinations arise in LLMs + show by targeted #intervention that these internal bias features do cause bad model outputs The trick: we get LLMs drunk and they tell us the

We're thrilled to finally share what we've been working on. Our new paper gives a first-ever glimpse into the "mind" of an LLM, and we discovered something that shocked us: AIs see ghosts in the machine. 👻 What do LLMs see in the dark?

Why do LLMs hallucinate? We may have the answer! We show that LLMs imagine MORE concepts in inputs with LESS semantic structure. Very surprising! Read the thread below for more details 👇 Praneet Sonia Luca Scimeca Danilo Bzdok