OpenAI znalazło 948 błędów w jądrze Linuksa, a banki blokują agentów AI

OpenAI wydało 1,2 mln dolarów na API, by znaleźć 948 luk w Linuksie. W tym samym czasie brytyjskie banki zamykają konta przed agentami AI.

Źródła:

• OpenAI Finds 948 Local Privilege Escalation Bugs In The Linux Kernel - Phoronix
• An unreleased Anthropic model just helped solve a major quantum-computing question
• Ripple-65B Preview delivers strong agentic coding results as a fully Apache 2.0 open US model
• Perplexity now has its own always-on AI agent
• "Most apps never make it off localhost": Cursor takes on Lovable and Replit with fully managed app deployment
• Banks ban AI from accounts over fears of rogue agents
• AI auditors catch most agent misbehavior but flood reviewers with false alarms, new study finds
• Nine AI agents tried autonomous research on open questions from US national labs and none passed the experts' bar
• State propaganda in some Chinese AI models is only "a thin layer" that fine-tuning can strip away
• AI coding agents still need human cleanup on nearly three out of four tasks in a real-world test
• Google Researchers Propose Sleep Framework to Retain LLM Facts
• OpenAI launches GPT-6 Cyber for gated offensive security research
• The gap between open and closed AI models in cyber capabilities keeps shrinking
• AI chatbots can flip-flop on false claims when users keep repeating them
• World Models Will Soon Be Inescapable. Here’s What You Need to Know

📰 Przeczytaj całe wydanie na stronie

🤖 Odcinek opracowany przy pomocy sztucznej inteligencji. Lektor syntetyczny.