Click to zoom
The recent uproar surrounding the Pentagon’s engagements with Anthropic’s Claude technology has left many startups keen on defense contracts scratching their heads. Just over a week ago, talks between the Pentagon and Anthropic fell apart, leading to the Trump administration labeling Anthropic as a supply-chain risk. In a bold move, the AI firm has decided to fight this decision in court.
OpenAI's Swift Reaction
At the same time, OpenAI wasted no time in securing its own agreement with the Pentagon, which sparked quite a backlash from users. The reaction was so intense that many opted to uninstall ChatGPT, causing Anthropic’s Claude to surge to the top of the charts in the App Store. This whirlwind even led to at least one OpenAI executive stepping down due to concerns that the announcement lacked necessary safeguards.
In the recent episode of TechCrunch’s Equity podcast, hosts Kirsten Korosec, Sean O’Kane, and I dissected the fallout for startups eager to collaborate with the federal government, especially the Pentagon. As Kirsten pointed out, “Are we going to see a changing of the tune a little bit?”
Pentagon's Scrutiny on AI
Sean shed light on this unusual scenario, especially given that both OpenAI and Anthropic have churned out highly publicized products—products that “no one can shut up about.” But it’s not just about AI capabilities; it’s also about the ethical implications of how these technologies might be harnessed in warfare, which naturally invites closer scrutiny from the public.
In light of these events, Kirsten speculated whether startups might rethink their strategies regarding government contracts. “After witnessing the standoff between the Pentagon and Anthropic, could other startups think twice about chasing federal funding?” she asked—and it’s a valid point.
The Defense Sector's Underbelly
Sean suggested, “I believe the immediate impact will be minimal because many firms, from fledgling startups to industry giants like General Motors, have historically kept a low profile while working with the Department of Defense.” GM, for instance, has a long-standing history of producing defense vehicles, including modern electric and autonomous models—yet this generally flies under the radar.
However, it’s worth noting that Anthropic and OpenAI are under an exceptional spotlight, as conversations around their technology touch on life-and-death matters. While OpenAI’s models might be widely discussed, General Motors’ defense contracts simply don’t ignite the same level of public discourse.
The Future for Defense Startups
So, will other companies like Applied Intuition shy away from dual-use projects? Sean doesn’t think so, as he believes that the complex judgments surrounding these partnerships often operate in a gray area.
Anthony emphasized how these events pose critical queries about technology and AI’s role in government operations. He remarked, “What's unique here is the Pentagon's aim to redo existing contract terms, which is rare and should definitely give startups pause.” The environment seems to be growing more unpredictable, particularly as traditional government procurement protocols face challenges.
The Fallout for OpenAI
OpenAI’s recent dealings have already begun to show adverse outcomes, with reports indicating a staggering 295% surge in uninstalls for ChatGPT following the DoD contract announcement.
The implications continue to unfold, but one pivotal question remains—the extent of unrestricted military access to AI models. The discussion on the TechCrunch podcast directly addressed what startups ought to consider when navigating federal contracts, especially in this uncertain climate shaped by U.S. policy concerning AI in military engagements.
Thanks for reading!
If you found this article helpful, share it with others