SELECT LANGUAGE BELOW

Anthropic resumes discussions with the Pentagon following CEO’s refusal to offer ‘dictator-like admiration’ to Trump.

Anthropic resumes discussions with the Pentagon following CEO's refusal to offer 'dictator-like admiration' to Trump.

Anthropic has reportedly re-engaged in urgent negotiations with the Pentagon to resolve ongoing disputes regarding AI safety measures. This comes just days after CEO Dario Amodei suggested that part of the issue arises from his rejection of what he calls “dictator-style praise” for former President Trump.

Amodei has been in discussions with Emil Michael, the undersecretary for research and engineering, in what is described as a “last-ditch effort” to formalize a contract outlining the military’s use of Anthropics’ AI models, according to a report from the Financial Times.

Negotiations resumed shortly after Amodei circulated a lengthy memo criticizing OpenAI for allegedly spreading falsehoods concerning its own conflict with the Pentagon, particularly regarding surveillance and autonomous weapons.

In a memo sent to employees, Amodei speculated that the administration’s antagonism may stem from their refusal to “donate to Mr. Trump.” He pointed out that OpenAI’s founder, Greg Brockman, has been a significant donor to the former president.

“The real reason the DoD and Trump administrators dislike us is that we are not part of that donor club,” he wrote, clearly expressing frustration.

Reaching an agreement would enable the Pentagon to keep utilizing Anthropic’s technology and could help the company avoid being labeled as a “supply chain risk.” Defense Secretary Pete Hegseth has hinted at potentially forcing companies in the military supply chain to sever ties with Anthropic.

Hegseth has yet to make that official designation.

Amodei, who previously backed Kamala Harris during her presidential run, has criticized both OpenAI and the Department of Defense for what he describes as misleading communication. He asserted that certain claims made by them were false, particularly regarding alleged terms from OpenAI that were supposedly offered to Anthropic and dismissed.

This memo followed shortly after OpenAI announced plans to provide AI services to the Defense Department following the breakdown of negotiations with Anthropic.

Sam Altman, OpenAI’s CEO, responded to the criticisms, stating that AI firms should prioritize directives from elected officials rather than allow the government to control policy.

He emphasized that it could be detrimental if companies neglected their democratic responsibilities simply because they disagreed with authority figures.

While standing by OpenAI’s contract with the Department of Defense, Altman noted that the timing appeared less than ideal, coming just after Anthropic was blacklisted, which he admitted “looked opportunistic and sloppy.”

All federal agencies have been ordered to halt the use of Anthropic’s services, and defense contractor Lockheed Martin has complied with this directive.

Investors have pressured Amodei to seek reconciliation with the Trump administration. He has also criticized OpenAI’s strategy concerning military safeguards, arguing that their contract offers insufficient protections while accepting weaker measures than those proposed by his company.

Amodei has accused Altman of undermining his company’s market position by securing Pentagon contracts, alleging that Altman is coordinating behind the scenes to replace Anthropic once it is officially categorized as a “supply chain risk.”

The breakdown in negotiations was primarily over language that Anthropic insisted should be included to prevent their AI from being used for domestic surveillance and fully autonomous weaponry.

Meanwhile, OpenAI is quickly modifying its contract language with the Department of Defense to incorporate extra safeguards, aiming to prevent the technology from being misused for spying on American citizens.

Modifications have already been made to prohibit any deliberate surveillance of U.S. citizens, and further protections are being discussed for implementation in the upcoming months.

These changes follow Anthropic’s refusal to accept similar contract terms, which led the Pentagon to opt for an agreement with OpenAI instead.

The Post reached out for comments from the White House, OpenAI, Anthropic, and the Department of the Army.

Facebook
Twitter
LinkedIn
Reddit
Telegram
WhatsApp

Related News