Emil Michael, now a senior technology official at the Department of Defense, has publicly aired long-held grievances over his exit from Uber and provided a detailed account of the government’s escalating dispute with AI firm Anthropic. The remarks, delivered in a recently released podcast interview, reveal a deep-seated bitterness toward Uber investors who ousted him and Travis Kalanick, as well as sharp criticism of Anthropic’s attempts to impose restrictions on its AI models used by the DoD.
The Unresolved Fallout From Uber
Michael, who resigned from Uber just days before Kalanick in 2017 amid a workplace investigation, described his departure as an “effective” ousting. While not directly implicated in the allegations of harassment and discrimination that triggered the investigation, Michael believes he was unfairly removed alongside Kalanick.
He harbors no forgiveness for the investors, particularly Benchmark, who pushed for their removal, arguing that their short-sighted focus on immediate profits killed Uber’s potential to become a trillion-dollar autonomous driving giant.
“I’ll never forget that, nor forgive,” Michael stated bluntly.
Kalanick shares this sentiment, lamenting that Uber’s self-driving program was on track to rival Waymo before its premature shutdown in 2020. He has since doubled down on robotics with his new venture, Atoms, and is poised to acquire autonomous vehicle startup Pronto, further signaling his continued commitment to the field.
The DoD’s Standoff with Anthropic
The interview also shed light on the DoD’s contentious negotiations with Anthropic, which ultimately collapsed. Michael argued that Anthropic seeks to layer its own policy preferences onto existing legal and internal regulations, creating an unacceptable restriction for military applications.
He further warned that Anthropic’s AI models have been repeatedly targeted by Chinese technology companies using a technique called distillation, effectively reverse-engineering their capabilities. This, according to Michael, would grant the People’s Liberation Army access to functionally equivalent AI technology while the DoD operates with a restricted version.
“I’d be one-armed, tied behind my back against an Anthropic model that’s fully capable — by an adversary,” Michael said, characterizing the situation as “totally Orwellian.”
Escalating Legal Battle
The dispute has since moved to court, with Defense Secretary Pete Hegseth deeming Anthropic a “supply-chain risk.” The government filed a 40-page brief arguing that granting Anthropic access to DoD infrastructure would introduce “unacceptable risk,” as the company could theoretically disable or alter its technology.
Anthropic fired back, submitting sworn declarations contesting the government’s claims, asserting that such interference is not technically possible. A hearing is scheduled for Tuesday in San Francisco to resolve the matter.
The ongoing clash between the DoD and Anthropic highlights the growing tension between national security concerns and the private sector’s control over cutting-edge AI technology. The outcome will likely set a precedent for future government partnerships with AI firms, shaping the landscape of defense innovation for years to come.
