Research scientists at Google DeepMind have stated they feel “deeply embarrassed” about a recent contract signed with the Pentagon, allowing the use of their AI technology in classified settings.
A senior researcher at Google DeepMind, Andreas Kirsch, voiced strong disapproval of the company’s new contract with the Pentagon in a post on X. He expressed, “We are speechless that Google has signed an agreement to use our AI models for sensitive work. Frankly, this is shameful.”
The contract was finalized despite considerable opposition from Google employees. Over 600 of them sent a letter to CEO Sundar Pichai, urging him to prevent the Pentagon from utilizing the company’s AI in sensitive areas.
Kirsch reflected on his initial hopes, stating, “When I went to bed yesterday, I was hopeful that my letter to employees would have an impact and give me time to consider it. I woke up this morning and Google had signed the worst version of the deal in the meantime.”
He expressed disappointment that Google proceeded without proper discussion. “It’s unfortunate that a major tech company is avoiding an honest conversation about this, especially when regulations are still catching up. It doesn’t cost that much, considering the overall size of the deal,” he added.
Researchers voiced personal struggles related to their roles within the company. “Honestly, I’m quite embarrassed to be a senior researcher at Google DeepMind right now, and I’m at a loss about how to function today,” he shared.
A Google spokesperson explained that the confidential labor deal is an amendment to an existing contract. Prior to this, the company held a contract with the Department of Defense for unclassified AI projects, due for completion in late 2024.
The spokesperson detailed Google’s commitment to supporting government agencies through classified and unclassified projects, noting areas like logistics and cybersecurity. They emphasized that Google is adhering to an agreement stating that AI should not be utilized for domestic surveillance or autonomous weapons without adequate human oversight.
Google’s actions reflect a widening trend in the tech industry, with more companies collaborating with defense agencies. OpenAI, Anthropic, and Elon Musk’s xAI are among those that have recently established ties with the Department of Defense. Some firms, like Anduril and Palantir, have designed their entire business plans around defense contracts.
In a social media statement, Kirsch contrasted Google’s deal with a similar classified agreement between OpenAI and the Department of Defense, questioning how OpenAI could sign a contract that seems to undermine trust.
The conflicts between conservative ideals and Silicon Valley firms that have thrived on anti-military sentiments are core themes of a new bestseller by Wynton Hall of Breitbart News.
Sen. Marsha Blackburn (R-Tenn.) recognized Hall’s book as a “must read,” emphasizing that it addresses the conservative struggle against Big Tech, while also considering how to harness AI’s potential responsibly.
Read more here.
