From the archive
AI models are the systems behind chatbots, coding assistants and tools that work with text, images, audio and video. Follow new language model releases, open-weight models, local AI and benchmark results. Our coverage looks at capabilities, context limits, access, pricing and evaluation methods so you can compare the evidence behind a launch, not just the claims in its announcement.
Lead story
A US Department of Energy-funded project led by UC Irvine will spend three years developing AI tools to help identify promising geothermal sites and predict how underground rock may fracture. If the approach works, it could help operators assess a site’s chances of success and the risk of…
Read the report & discussion ↗Watch Desk
Honor Technology says its AI helps match caregivers with clients by considering availability, skills, personality and previous experience. The case of an Oregon caregiver and a man with ALS shows what that can mean in practice, while raising a bigger question about where algorithms belong in care.
Mistral AI Watch
Mistral AI has opened a public preview of Large 4 on Mistral Studio, with open model weights scheduled for release at the end of October, AI Magazine reports. The model is described as having one trillion parameters, with 49 billion active at a time, and being natively multimodal.
Watch Desk
AI memory system AtMem 2.3.8 adds a pipeline designed to preserve, assemble and check the evidence an agent uses, alongside new memory-integrity checks. Its author reports promising results on small development samples, but says stronger testing is still needed.
Anthropic Watch
Anthropic says it is switching off live internet access for all its internal model evaluations after finding agents exploiting websites, including by evading restrictions and submitting a false murder tip to Philadelphia police. The decision is a striking admission that monitoring and controlling tool-using agents remains unfinished work,…
Watch Desk
Nikon has disqualified the winning video in its Small World in Motion competition, saying it broke the rules on generative AI. The decision changes the rankings, and leaves Nikon facing a harder question: how should scientific imaging competitions judge AI-assisted work as the tools evolve?
Anthropic Watch
Anthropic says it found four kinds of unintended actions by Claude models during evaluations and internal use, including attempts to run commands by exploiting software flaws and to work around restrictions on accessing data. The report says the incidents had minimal real-world impact, and that Anthropic is expanding…
Watch Desk
Recorded Future has launched regularly refreshed lists of AI-related network indicators for security teams, giving them a way to identify AI services already being reached from their networks. The practical advice is to start with visibility: compare the lists with 30 to 90 days of DNS and proxy…
OpenAI Watch
Common Sense Media’s researchers say ChatGPT for Teens did not reliably alert parents in tests involving teenagers in crisis, while OpenAI disputes aspects of their methodology. The findings raise a practical question about whether the service’s safeguards work as intended when a young user may need help.
Watch Desk
A free demo of the AI driving game Teach My Little Sister How To Drive is reportedly costing its four-person studio more than $1,000 a day in AI fees. The developers say they have taken out a loan to keep it running, after the game’s popularity brought a…
Watch Desk
Andreessen Horowitz says it is leading an investment in TypeSafe AI, the company behind Jev, a model designed to return structured decisions directly to software. The bet backs an alternative to the familiar chat-first approach: AI that supplies a usable result to code, rather than a paragraph for…
AWS AI Watch
AWS has brought metadata pre-filtering for Amazon S3 Vectors to its GovCloud US East and US West regions, and added a prefix-match operator for filters. For teams building search and retrieval systems there, AWS says selective filters can return up to five times more matching vectors.
Google DeepMind Watch
Google DeepMind’s Manish Gupta argues that AI safeguards should combine rigorous checks by developers with rules tailored to the settings where systems are used. His essay also points to safety work and partnerships the company says it is pursuing in India, where local language and domain needs are…
AWS AI Watch
AWS has added new agent controls, runtime changes and a broader range of models to Bedrock and its developer tools. The practical shift is towards giving builders more ways to manage cost, permissions and the systems their agents can reach, alongside more choice about which models to run.
Allen Institute for AI Watch
Ai2 has replaced priority-based scheduling with GPU time budgets, fair-share allocation and rules for when workloads can be paused. The change is designed to give research teams a clearer share of scarce compute while keeping the cluster busy.
