The White House’s plan to vet potentially dangerous AI is cloaked in secrecy
TL;DR
After months of talks with tech industry leaders, the Trump administration has finalized a framework for testing new AI models for safety and cybersecurity risks. Staff from OpenAI, Anthropic, Meta, Google, Nvidia and Microsoft reviewed it in a private meeting at the White House. The volunteer vetting process is settled, but the administration does not plan to publish the policy and will share testing criteria only with a select few companies.
Nauti's Take
Having a shared testing framework for safety and cyber risks at all is progress compared with having no common yardstick. The problem is the secrecy: if only a handful of companies know the criteria, nobody outside can check or compare how strictly models are actually tested.
The large labs gain from sitting close to the process, while smaller providers and researchers are asked to take the results on faith.
Sources