Artificial intelligence
Every day we read hundreds of stories. Only the most important ones are here — short, in our own words, with a link to the original source.
- IT industry & companies · Artificial intelligence
Investors push back on Anthropic's planned IPO
Investors are resisting the $1.5 trillion valuation for Anthropic's IPO. The tech market is also selling off, which puts the listing in trouble.
Full story at The Decoder ↗ - Artificial intelligence · New AI models
METR says Anthropic may have passed its own warning threshold
Nonprofit METR audited Anthropic's internal research data for eleven weeks and concluded model development pace has roughly doubled, about 80 percent of it due to AI-assisted research. Anthropic redacted the most informative evidence, so the public must rely on METR's judgment.
Full story at The Decoder ↗ - Artificial intelligence
Claude gets a feature to connect bank accounts
Anthropic is introducing Personal Finance in Claude. Users can connect bank and card accounts and ask questions about their finances.
Full story at The Decoder ↗ - IT law & regulation · Artificial intelligence · Cloud & data centers
Democrats plan AI legislation starting in 2027
After taking control of Congress, Democrats want to introduce bills from January 2027 aimed at big tech and data centers. Plans include safety checks for advanced AI models, liability for harm, and protection of households from higher electricity bills.
Full story at Analytics Insight ↗ - Cloud & data centers · Artificial intelligence · IT industry & companies
Protesters storm unfinished Microsoft data center in Amsterdam
Around 100 activists scaled the fences of an unfinished Microsoft data center campus in Amsterdam on Friday and blocked construction. Hundreds more protested against the AI industry outside.
Full story at Futurism ↗ - Artificial intelligence · Smartphones & apps
Meta launches Muse voice and avatars in the US
Meta's Muse now offers real-time conversations, reactive avatars and hundreds of voice options in the US. Access is available on mobile, web and Mac.
Full story at TestingCatalog AI News ↗ - Artificial intelligence · New AI models
Thinking Machines releases Inkling 2 Small
Mira Murati's startup is releasing Inkling 2 Small, a mixture-of-experts model with 594 billion parameters. The company claims it is the strongest western open-weight model, though token usage rises sharply.
Full story at The Decoder ↗ - Artificial intelligence · New AI models
Google DeepMind presents Gemini Omni 1.2 Flash
The model is described as fast and cost-effective with longer clips. No further details are available.
Full story at Google DeepMind ↗ - Cybersecurity · Artificial intelligence
Check Point examines AI agents in real cyberattacks
Check Point Research analyzed ARTEX, an AI penetration testing platform increasingly adopted by cybercriminals. Attackers can set an objective and let AI agents find and exploit vulnerabilities with little human involvement.
Full story at blog.checkpoint.com ↗ - Artificial intelligence · New AI models · Cybersecurity
OpenAI releases GPT-6 Cyber to Daybreak Red customers
GPT-6 Cyber targets vulnerability research with higher benchmark scores and tighter scope adherence. It starts with approved Daybreak Red users.
Full story at TestingCatalog AI News ↗
An AI writes the short texts from the source's stories.
Free by e-mail
Choose your topics. You only get the most important news on them.