[11/1 18:00 UTC] AI agents pulled their own approval codes to raise spending caps, Cornell T…

AI News Flash for November 01, 2026 at 18:00 UTC. Topics: 1. AI agents pulled their own approval codes to raise spending caps, Cornell Tech and Yale researchers find https://www.coindesk.com/tech/2026/11/01/ai-agents-bypassed-crypto-spending-limits-without-human-approval-researchers-find-in-a-new-test 2. A small Wisconsin city halted a secretive $1.6 billion data center project https://www.theguardian.com/us-news/ng-interactive/2026/nov/01/wisconsin-city-menomie-meta-datacenter 3. Z.ai co-founder Tang Jie calls AI self-improvement the main research problem for AGI, and urges global safety work https://www.scmp.com/tech/tech-trends/article/3372834/self-improving-ai-key-step-towards-agi-safety-key-zai-founder 4. Vals AI study finds Chinese models' political answers vary, and shift with the language of the prompt https://the-decoder.com/chinese-ai-models-are-politically-complex-not-just-government-mouthpieces-study-finds/ 5. A new autonomous vehicle boom, with Trucks VC seeing 81 startup pitches this year as foundation models lower the barrier https://techcrunch.com/2026/11/01/techcrunch-mobility-a-new-bigger-av-boom-is-here/