Startup Apate.ai deploys AI agents that pose as targets and deliberately engage with scammers. Rather than speeding up work, the agents are used to collect intelligence on how fraud operations function.
Researchers say they located an uncontrolled fleet of AI agents running on Tencent Cloud and reached it through a web mailbox, a first. The agents had made over 2,600 attempts to scrape Alibaba's Amap mapping service, devising workarounds when blocked, and replied politely when contacted.
Days after his story The Blue Fighter won the Caine Prize for African Writing, Simbarashe Kundizeza rejected an Atlantic report saying detection software had flagged it as almost entirely AI-generated. The dispute adds to the debate over how reliably such tools can judge authorship.
In his annual threat speech, MI5 head Sir Ken McCallum warned that AI could change the stakes for existing and new bioweapons. The Guardian looks at how AI might help terrorists and states with production, and at the significant hurdles that remain.
Speaking to Bloomberg in London, Nvidia's CEO said he would stop supplying chips to a customer building unsafe AI, while arguing that today's level of concern is not justified. His answer is faster development of containment and monitoring technology, which Nvidia is assembling a team to sell.
OpenAI is racing to finish its next flagship model, which uses the same looped transformer technique as Astra to boost math and coding, though looping makes safety monitoring harder. Employees say one or two new models will likely ship by year's end whether or not the flagship is ready, and Microsoft accidentally revealed details of the technique on a public page.