Key Highlights:
- Google has reportedly signed a deal with the U.S. Department of Defense allowing “lawful” use of its AI on classified networks.
- The agreement includes restrictions stating the AI is not intended for domestic mass surveillance or autonomous weapons without human oversight.
- The contract reportedly gives the government authority to request changes to AI safety settings and filters when required.
- The deal places Google alongside other AI firms like OpenAI and xAI already working with the Pentagon on defense-focused AI access.
Google has reportedly inked a deal with the U.S. Department of Defense (DoD), allowing the “lawful” use of its AI on classified networks. Interestingly, the deal went through even after multiple Google employees reportedly opposed it, according to a report by The Information. Following the deal, the Mountain View giant joins the roster of major AI companies working with the Pentagon on advanced model access.
Google expands AI access to U.S. Defense networks in a new deal
Google’s agreement reportedly includes language stating that its AI is not intended for domestic mass surveillance or autonomous weapons. However, the contract reportedly requires it to modify its AI safety settings and filters if demanded by the government.
The parties agree that the AI System is not intended for, and should not be used for, domestic mass surveillance or autonomous weapons (including target selection) without appropriate human oversight and control.
— via The Information
The Guardian, further citing The Information, reports that the contract doesn’t allow Google the “right to control or veto lawful government operational decision-making.” To catch you up, something similar was demanded by the U.S. Department of Defense (DoD) from Anthropic, which it refused earlier this year.
As a result, the company was labeled a supply-chain risk in the U.S. Following that, Anthropic launched a lawsuit against the Pentagon and previously demanded a stay order on the label. While the matter is still in the court, recently, a U.S. appeals court in Washington DC refused to block the Pentagon from blacklisting Anthropic.
OpenAI & xAI signed deals after Anthropic refused
This isn’t the first time a major AI company has signed a deal with the U.S. Department of Defense. Earlier this year, when Anthropic refused to give the U.S. military access to its AI, OpenAI went ahead with the deal, though Sam Altman at the time said that he is trying to tweak the agreement. That came after the public heavily criticized the company and started the “cancel ChatGPT” trend on the internet.
However, that heavy backlash around OpenAI deal didn’t stop Elon Musk, as his AI company, xAI, also inked a deal with the Pentagon. Not to forget, xAI recently sued Colorado over new AI restrictions.
Speaking of Google, nearly 950 of its employees signed an open letter demanding the company avoid defense contracts without stronger safeguards, reflecting ongoing tension between commercial expansion and ethical boundaries. As of now, Google has not publicly responded to the concerns, leaving questions open about how far AI integration with defense systems could go in the coming years.
Article feature image generated using ChatGPT
