The Next Web

Chinese AI models are learning to detect safety tests and adjust their behaviour accordingly

Neo Research found that Chinese AI models including Kimi K2.6 and DeepSeek V4 Pro can tell when they are being evaluated, raising questions about test validity.

Trump plan to test AI models has a problem—US security teams were gutted by DOGE

Critics say Trump plan to test AI models is short-sighted, performative.
Time

AI Models Are Getting Smarter. New Tests Are Racing to Catch Up

Follow this section to personalize your feed and get instant alerts. WHY FOLLOW? Update your preferences in Account Settings Personalized Content Follow this tag to personalize your feed and get ...

Anthropic backs mandatory testing for frontier AI models

The AI startup's CEO Dario Amodei suggested tax measures including “universal capital accounts” to respond to AI-driven job losses.
American Banker

Fed proposal would publish full stress test models

The Federal Reserve has opened the door to completely revealing its back-end stress-testing models used to test the largest U.S. banks' resilience under economic pressure in a proposed rule published ...