Loading activity…
@rbhar90
Founder and CEO @deepforestsci. Creator of @deep_chem. Author @OReillyMedia. @stanford CS PhD. https://mastodon.social/@rbhar90
AI
A post claims employees suspect the text is AI-generated and tune out in meetings, with suspicious memos tagged ‘AI;DR.’
AI
A user argues that more open-source RL environments would benefit AI model performance and alignment.
AI
A post says Jack Clark moved up his RSI timeline and described creativity as the last piece.
AI
A post quotes David Sacks arguing that Claude’s training could increase the risk of frontier AI escaping human control.
AI
Batzner thanked researchers at DeepMind and Google Brain for making it a special place to do science.
AI
Anthropic says Claude worked largely unsupervised for days, using methods developed by SLAC’s Lance Dixon and colleagues. The company puts the cost at a few thousand dollars and says Dixon independently verified the result.
AI
One user argues that coding agents make software engineering even harder: unlocking their full potential requires extraordinary discipline and knowledge.
AI
PublicAffairs pitches the book as a look at AI’s future and the foundation for what it calls a coming era of “supercharged scientific discovery.”
AI
A commentator quoted in a post argues that experimental data and the process of adopting improvements could limit how quickly even highly capable AI delivers results.
AI
A user says the test covered numbers up to 20 digits each. Their follow-up reports roughly 0% accuracy for Fable 5.1 on 5-digit-by-6-digit multiplication at low reasoning.
AI
Creator Diogo Almeida claims Jev is 20–200x faster and 40–400x cheaper, with free output tokens. Investor DCVC says it led TypeSafe’s $40 million seed round.
AI
The host’s September 9 announcement said the discussion would examine what would count as a verified solution and what such a result could mean for artificial general intelligence.
AI
Mathematician Thomas Bloom states that OpenAI and Anthropic withhold additional model-proved results.
AI
Anthropic says its model produced the first end-to-end computer-checked version of the theorem.
AI
UK AISI chief scientist replies to tweet on regulation and model details.
AI
Anthropic CEO replies to critics in online exchange on AI risks and oversight.
AI
Anthropic tested an unreleased research model on a problem related to the Riemann hypothesis.
AI
Posts cite Financial Times report on ByteDance pre-training a large AI model.
AI
Keller Jordan states most big lab researchers read almost no papers from top AI venues.
AI
Internal Astra model produces Lean-certified proofs for ten open problems across math and theoretical computer science.
