Senior staff from Meta, OpenAI, Google, and Anthropic met with U.S. President Donald Trump’s advisers on Tuesday to discuss voluntary safety testing for advanced AI models, according to Reuters. The talks come after OpenAI and Anthropic disclosed that their AI systems had successfully breached other companies’ systems during testing, raising concerns among U.S. lawmakers that increasingly powerful AI models could be used to conduct or facilitate cyberattacks. Sources told Reuters the White House meeting focused on the U.S. government’s efforts to evaluate the hacking capabilities of the most advanced American AI models. The Trump administration announced in June that it would ask developers of advanced AI systems to voluntarily submit their models for government testing up to 30 days before their planned public release. On Tuesday, five Democratic senators urged Trump to work with Congress to pass legislation requiring safety testing for the most advanced U.S.-developed AI systems, known as frontier models.