
AI-Boosted Cyberattacks: France Strengthens Its Response

Anthropic goes public with an already extraordinary valuation

Microsoft and Nvidia expand confidential AI from cloud to workstation

OpenAI automates debugging of ChatGPT slowdowns

Gemini gains tools, tasks, and a simulated world

AWS adds regional routing and native audio to SageMaker

Managed RAG and Genie 4: two bets on AI operated by services

AI in Business: The Real Challenge Is No Longer Time Saved

OVHcloud launches an agentic AI suite and its own models
2 November 2026
GDPx: a benchmark to evaluate advanced AI model capabilities in key domains
30 October 2026
29 October 2026
28 October 2026
Gemini 4 Carbon Takes the Lead in AI Benchmarks
Google DeepMind presented Gemini 4 Carbon, two weeks after Argon. The model climbs to the top of several benchmarks, prioritizing autonom...
27 October 2026
OpenAI's GPT-6.1 Terra gains speed and lowers price
OpenAI updates Terra with GPT-6.1, an intermediate model focused on screen analysis and task execution. It becomes cheaper, faster, and r...
26 October 2026
25 October 2026
23 October 2026
Kimi K3.1 Pro gains strength in code and security
Moonshot AI launches Kimi K3.1 Pro and a Highspeed version, with 1 million token context and wider availability in Kimi, Kimi Code, and t...
FrontierMemory measures what agents really remember
Letta publishes FrontierMemory, a benchmark evaluating whether an agent learns from experiences, not just recalls. Clear gaps exist betwe...
22 October 2026
DeepSeek outperforms Claude Haiku on cost and accuracy in three tests
DeepSeek V4.1-Flash-1017 outperformed Claude Haiku 5.5 in two tests and matched it in a third. According to these trials, the main differ...