Sam Altman admits rushed OpenAI’s deal with Pentagon ‘looked opportunistic, sloppy’; issues clarification after backlash

Amid widespread user outcry, CEO Sam Altman acknowledged in an online clarification post that the company’s rush to sign a deal with the Pentagon “seemed opportunistic and sloppy.”

A significant number of Americans canceled their OpenAI and ChatGPT subscriptions to switch to other alternatives after Altman announced on Saturday that the company had reached an agreement with the US Department of Defense to deploy its models in secret networks.

The announcement also comes at a time when the US government is in a public spat with rival Anthropic PBC over “full military use” of AI models and insisting on some limitations. The two main sticking points are the use of AI for fully autonomous weapons and domestic mass surveillance.

Anthropic’s main app has soared to the top of the download charts, a sign of Apple’s support for the company during its conflict with the Pentagon. Altman’s newest post is seen as damage control. Here’s a look at the key highlights:

‘Consistent with applicable law, critical to the protection of civil liberties’: Altman

In a post on social media platform

• At the time of writing, the post had reached one million views within a few hours. The OpenAI chief said the agreement would be amended by adding language stating that it was “consistent with applicable law, including the Fourth Amendment to the United States Constitution, the National Security Act of 1947, the FISA Act of 1978.”
• He added that it will be stated that the AI ​​system will not be used intentionally for domestic surveillance of US citizens and citizens.
• “For the avoidance of doubt, the Department understands that this limitation is meant to prohibit intentional tracking, surveillance, or monitoring of U.S. citizens or nationals, including the purchase or use of commercially obtained personal or identifiable information,” he added.

Notably, a source told Axios last month that the government rejected Anthropic’s requests because there was “a significant amount of gray area as to what would and wouldn’t fall within” the categories in dispute and because the Pentagon was not willing to negotiate each case on a case-by-case basis or let Anthropic’s models unexpectedly hinder some processes.

Altman said protecting Americans’ civil liberties is “critical” and that the company “wants to specifically clarify this point, including with commercially obtained information.”

• “To put it very clearly: We want to work on democratic processes. The government should make the basic decisions about society. We want to have a voice and a seat at the table where we can share our expertise and fight for the principles of freedom,” he said.
• Altman also said, “If he gets an order that I believe is unconstitutional, of course I would rather go to jail than comply with it.”

‘There are a lot of things that technology is not ready for’

The OpenAI chief also added that there are “a lot of things that AI is not ready for, and a lot of areas where we don’t yet understand the trade-offs needed for security,” and the department said they will “slowly work on those” with technical security measures and other methods.

In a rare admission, he added: “There’s one thing I think I did wrong: We shouldn’t have rushed to announce this on Friday. The issues are extremely complex and require clear communication. We were really trying to deescalate things and avoid a much worse outcome, but I think it just seemed opportunistic and sloppy. It’s a good learning experience for me as we face more important decisions in the future.”

He also reiterated his support for Anthropic and said it should not be identified as a supply chain risk (SCR), adding: “We hope DoW will offer them the same terms we agreed on.” He first made this sentiment known Sunday following the Pentagon’s action against the company led by CEO Dario Amodei.

In particular, the leaders of the ChatGPT maker and the Claude AI maker have clashed many times in the past over different approaches to AI development.

Sam Altman summarizes the principles: Harmony, democratization, empowerment and individual agency

In the same title, Altman called the deal with the Pentagon “one of the first ‘real deal’ decisions we’ve ever encountered,” and in doing so, he shared the principles he holds most dear: harmony, democratization, empowerment, and individual agency.