Loading activity…
@schwarzjn_
Prof @imperialcollege, Research @oiioxford | ex- @Harvard, @GoogleDeepMind, exited Founder, @Reuters | PhD @gatsbyucl 🇬🇧🇩🇪🇭🇰
AI
Author Nathan Benaich cites an Anthropic index and argues robotics is nearing a “GPT-2 moment.”
AI
A user thanked its writer directly, calling it “such an important book.”
AI
Mistral says Mistral Large 4 is live via API in public preview now, with open weights due at the end of October. Reuters says the launch is part of Mistral’s push to narrow the gap with leading Chinese open-weight models.
AI
Aleph Alpha says the bilingual mixture-of-experts model has 78.1 billion parameters and a context option up to 1 million tokens.
AI
Posts on X congratulated Nathan Lambert and Trillium Labs after a launch, with several users describing the effort as open, science-oriented or focused on open post-training. Lambert later said some criticism came from Substack and more general-audience AI followers, not the technical AI community.
AI
A user urges young AI researchers in London to found or join startups there, arguing they could have more impact.
AI
Anthropic reports GLM-5.3 built end-to-end exploits in 50 of 410 offline test attempts, versus 56 for Mythos Preview.
AI
Palisade says it interviewed 22 current and former employees from OpenAI, DeepMind and Anthropic about their personal views and fears around AI development.
AI
A user recommends the course as an Oxford-taught technical foundation, pairing the pitch with a secondhand claim about $650,000-a-year pay at Anthropic.
AI
A Cambridge visitor contrasts AI discussions indoors with cows relaxing in the shade outside, urging people not to neglect their mental health.
AI
FOX4 reports that President Donald Trump outlined the terminology change at the United Nations General Assembly, saying it would apply to all U.S. documents.
AI
A post summarizing Mozilla's 91-page report says eight of OpenRouter's 10 most-used models by August token volume were open-weight. It also cites open-model use by 79% of surveyed developers.
AI
Amodei says Anthropic will give third-party evaluators permanent, employee-level access to its systems to verify adherence to safety measures, report incidents and assess models’ alignment during training.
AI
The user says the approach tracks a reasoning problem’s unresolved cognitive demand and integrates it into a hierarchical meta-reasoning system, producing large gains for several frontier-scale models.
AI
His post reflects on building a team of 50+ researchers at a startup and early sovereign AI work.
AI
Researcher shares screenshot of Mistral Large 3 output on Russia via public AI constitution site.
AI
AI researcher Jonathan Richard Schwarz shares a screenshot of Grok 4.3 output.
AI
AI researcher Jonathan Richard Schwarz shares a Public AI Constitution webpage.
AI
Jonathan Richard Schwarz shares Kimi K3 screenshot on Chechen Wars with stark heading.
AI
The project says explicit priorities can help a model navigate conflicts between honesty, helpfulness and harm avoidance. Its website describes the constitution as open for community review.
