[AI Companies Agree to Grant Early Access to the U.S. Government for Model Safety and Performance Evaluation] DeepTech TechFlow reports that on May 5, Alphabet's Google, Microsoft, and xAI have agreed to grant the U.S. government early access to their artificial intelligence models to evaluate system capabilities and enhance system safety before the technology is released to the public. According to a statement released by the U.S. Department of Commerce on Tuesday, through these agreements, AI developers will collaborate with OpenAI and Anthropic PBC to allow the Department of Commerce's AI Standards and Innovation Center to conduct pre-release reviews of their models. The agency stated that OpenAI and Anthropic have renegotiated their existing partnerships with the center to better align with the priorities outlined in Trump's Artificial Intelligence Action Plan. (Jin10)