Ultraman admits signing a contract with the US military was "too reckless"; OpenAI hurriedly revised the contract, strictly prohibiting the use of its models for domestic surveillance.

```

The AI contract signed between OpenAI and the US Department of Defense was revised within just a few days. CEO Sam Altman publicly admitted that the rushed signing “seems opportunistic and hasty,” and announced the addition of new terms, explicitly prohibiting the use of its AI systems to surveil American citizens.

According to the UK Financial Times, Altman stated on Monday that the company is negotiating supplementary clauses with the Department of Defense to ensure that “AI systems are not intentionally used for domestic surveillance of US citizens and nationals.” Intelligence agencies such as the NSA will also be temporarily excluded from the contract.

The contract, signed last Friday, authorizes the Pentagon to use OpenAI models in confidential military operations. However, after the contract was signed, internal company employees immediately expressed dissatisfaction, and graffiti protests reading "Oppose Mass Surveillance" appeared outside the San Francisco office.

This revision not only reflects OpenAI’s internal and external pressures regarding military collaboration, but also prompts a renewed examination of the company’s fundamental differences with its competitor Anthropic in handling similar issues—Anthropic’s negotiations broke down because it held firm on its red lines, and afterward suffered contract retaliation from the Trump administration.

Rushed Signing Sparks Backlash, Altman Admits Fault Publicly

The contract between OpenAI and the Pentagon was signed last Friday, precisely hours after Anthropic’s negotiations with Defense Secretary Pete Hegseth broke down. At the time, OpenAI claimed the agreement had “the strictest safeguards of any confidential AI deployment agreement to date, including terms superior to Anthropic’s.”

However, this statement quickly triggered dual internal and external pressure. According to sources, OpenAI employees expressed concerns through internal channels and spoke out publicly on social media. Over the weekend, chalk graffiti appeared outside the San Francisco office with “Oppose Mass Surveillance” written, urging employees to “do the right thing.”

Altman posted on social media Monday admitting: “We should not have rushed the contract out on Friday. These are extremely complex issues that require clear communication. We genuinely wanted to ease tensions and prevent worse outcomes, but I think it did indeed come across as opportunistic and careless.”

The supplementary terms announced on Monday will “prohibit intentional tracking, surveillance, or monitoring of US citizens or nationals, including through the acquisition or use of commercially obtained personal or identifiable information.” OpenAI also stated it would employ technical measures to prevent model abuse, including deployment only via cloud and not directly on military hardware, with company employees supervising the entire process.

Anthropic Negotiations Collapse, Trump Administration Pressures

OpenAI’s contract revision also draws attention to the breakdown of Anthropic’s negotiations with the Pentagon. According to a person familiar with the negotiations, Anthropic CEO Dario Amodei drew two red lines: prohibiting the company’s AI models from being used for domestic mass surveillance and prohibiting their use in lethal autonomous weapons.

Hegseth demanded the models be available for “all lawful uses,” but Anthropic executives believed that current US law already allows mass surveillance using AI tools, so insisted that explicit safeguards must be written into the contract before new legislation is passed. Last Friday, the Pentagon signaled a willingness to remove ambiguous wording Anthropic was concerned about, and top executives believed an agreement was near. However, negotiations stalled over terms related to large-scale collection of public data and ultimately collapsed.

After the talks fell through, the Trump administration immediately retaliated. The US Treasury Department, Federal Housing Finance Agency, and government-backed mortgage giants Fannie Mae and Freddie Mac all announced Monday they would terminate their contracts with Anthropic.

Risk Warning and DisclaimerThe market has risks, and investment must be cautious. This article does not constitute personal investment advice, nor has it taken into account individual users’ unique investment goals, financial situations, or needs. Users should consider whether any opinions, viewpoints, or conclusions in this article are suitable for their specific circumstances. Invest at your own risk. ```