“[W]Shouldn’t have been rushed to publish this on Friday,” Altman wrote in an X post on Monday. “The issues are extremely complex, and demand clear communication. We were really trying to minimize things and avoid a very bad outcome, but I think it just seemed opportunistic and sloppy.”
OpenAI CEO Sam Altman reacts to dealing with War Department
OpenAI announced its partnership with DOW late last week, terminating the contract within days of President Donald Trump ordering federal agencies to stop using competitor Anthropic. According to Anthropic CEO Dario Amodei, the split occurred because it rejected DOW’s demand that it remove safeguards against the use of AI for mass domestic surveillance and fully autonomous weapons. Instead, DOW wanted to use Anthropic’s AI tools for “any lawful use.”
As such, OpenAI’s accelerated DOW deal provoked immediate reaction from its citizen users. Despite OpenAI’s claims that its deal has even more safeguards than Anthropic’s original agreement, the contract allows both mass surveillance and AI-controlled weapons as long as such use is legal, and even sets out the circumstances under which it would be allowed.
Now OpenAI is attempting damage control, saying it has worked with DOW to add new language to the contract that directly addresses the use of its technology for home surveillance.
“Throughout our discussions, the Department [of War] “It has been made clear that it shares our commitment to ensuring that our devices will not be used for domestic surveillance,” OpenAI wrote in an update to its original deal announcement on Monday.
OpenAI updates War Department deal after backlash
Unfortunately, the new amendments that OpenAI shared continue to rely on legality as a preventive boundary preventing mass surveillance, leaving the possibility of such use open if the US government changes the law. They also failed to address the issue of autonomous weapons.
mashable light speed
The new sections read, “Consistent with applicable laws… AI systems will not be used to knowingly conduct domestic surveillance of U.S. persons and citizens.” “For the avoidance of doubt, the Department considers this limitation to prohibit the intentional tracking, monitoring, or surveillance of U.S. persons or citizens, including the procurement or use of commercially obtained personal or identifiable information.”
This tweet is currently unavailable. It may be loading or may have been removed.
Many social media users reacted with skepticism to OpenAI’s contract changes, with some arguing that its specific ban on “intentional” surveillance leaves notable loopholes.
Political researcher Tyson Brody (@tysonbrody) responded to Altman’s post, “It’s not hard to accept an AI dragnet.” “‘Intentionally’ and ‘intentionally’ – so American Desire This data may be tampered with, but the government can claim ‘incidental collection’ and thus it is legal.”
@Andy_Bloch wrote, “‘Not intentionally used’ is not real safety in an autonomous AI system.” “It may stop monitoring because of what it was trained on, what it detects, or how people use it later.”
Altman had previously indicated that OpenAI would limit the use of its AI tools only on legal grounds, not ethical grounds, during a Q&A held shortly after the DOW deal was announced. The CEO expressed reluctance to take an ethical stance, saying that OpenAI prefers to follow government directives rather than consider such issues.
Despite criticism of this apparent abdication of responsibility, Altman reiterated this position in his new memorandum, calling it respect for “democratic processes”.
“Government must make important decisions about society,” Altman wrote. “We want a voice, and a seat at the table where we can share our expertise, and fight for the principles of freedom. But we are clear about how the system works (as many people have asked, if I get what I believe is an unconstitutional order, of course I would rather go to jail than comply with it).”
Altman said that Dow intelligence agencies such as the National Security Agency (NSA) would not use OpenAI’s technology without amending their contracts. Nevertheless, it currently seems unlikely that OpenAI would refuse legal requests for such modifications, no matter what ethical issues might arise. (The NSA was previously revealed to be conducting mass surveillance of American citizens by whistleblower Edward Snowden in 2013.)
Uninstalls reportedly increased by 295 percent in the wake of the news that many OpenAI customers had canceled their ChatGPT subscriptions in response to the company’s deal with DOW. Anthropic’s AI Chatbot Cloud has overtaken ChatGPT as the most downloaded free app in the US Apple App Store.
Disclosure: Mashable’s parent company Ziff Davis filed a lawsuit against OpenAI in April 2025, alleging it infringed Ziff Davis copyrights in the training and operation of its AI systems.
Subject
Artificial Intelligence OpenAI
<a href