Microsoft (MSFT), Google (GOOGL), and xAI, the firm founded by Elon Musk, have agreed to grant the U.S. government early access to their latest artificial intelligence (AI) models. The move comes as the AI and tech firms sign a new partnership with the government.
Claim 55% Off TipRanks
Forget margin or options. Here's how the pros trade AMZNThe deal will assist the authorities in assessing the potential risks linked to these models before their public release. The agreement was revealed by the Center for AI Standards and Innovation (CAISI), which operates under the U.S. Department of Commerce. The move shows that the government is making firm efforts to stay ahead in the fast-growing AI space and is watching how it may affect national security.
Early Testing Framework Expands Government Oversight of AI
The new deal with the U.S. government allows the CAISI to test AI models built by firms to better learn their capabilities and find loopholes that could be exploited. The center has already carried out more than 40 tests on advanced models that have not yet been made public.
Under the new deal, xAI, Microsoft, and Google will have to submit the unguarded versions of their AI models. This allows the agency to conduct deeper, more thorough checks into how the models behave without restrictions. It also helps to reveal issues that might have gone unnoticed.
CAISI director Chris Fall emphasized how crucial these strict evaluations are in building safe and secure AI systems. He noted that they are critical for checking how AI tools can affect national security before they are widely used.
Notably, this new agreement is an expansion of an earlier one with OpenAI and Anthropic, signed in 2024. At the time, the CAISI operated under the name U.S. Artificial Intelligence Safety Institute (AISI).
Security Concerns and Pentagon Deal Drive Collaboration
Interestingly, the new deal comes at a time when the safety concerns around AI-related risks are high. Recently, the Pentagon’s CTO, Emil Michael, found potential risks related to Mythos, the new AI model created by Anthropic, the firm behind Claude AI. Experts say the new model can find software flaws faster than most humans.
Around the same period, the Department of Defense (DoD) unveiled a deal with some major AI and tech firms, excluding Anthropic. The deal will allow these firms to work on the military’s most classified systems and deploy their AI tools within them.
The deal also allows the U.S. government to access advanced AI models for its operations without relying on a single provider. This gives the government room to improve in areas like planning, logistics, and cybersecurity.
At the same time, Kent Walker, President of global affairs at Google, confirmed the firm’s plan to support the defense agencies. He noted that the company’s interests align with those of other leading AI developers working with the government to improve national security.
What Are the 5 Best AI Stocks to Buy Right Now?
Wall Street analysts rate Nvidia (NVDA), Microsoft (MSFT), Meta Platforms (META), Amazon (AMZN), and Google (GOOGL) as Strong Buy, based on TipRanks consensus data. Among these stocks, NVDA has the highest upside potential of 36.05%, with an average price target of $274.38. For more information on the performance, rating, and price target of each of these stocks, visit the TipRanks Stocks Comparison Center.
