Technology 5 min read By Alice Ashford
OpenAI Human Rights Lead Warns Military AI Use 'Keeps Me Up at Night'
Sarah Yager, OpenAI's first Human Rights and Responsible Deployment Lead, has voiced deep concerns about military applications of artificial intelligence, particularly in targeting and autonomous weapons, while defending her decision to work inside the company.
Sarah Yager, OpenAI's first Human Rights and Responsible Deployment Lead, has said that the military's use of artificial intelligence in targeting and autonomous weapons is a source of profound personal concern. Speaking on stage at the ScaleUp:AI conference in New York City, Yager said the question of what armed forces will do with AI 'keeps me up at night', adding that 'everyone is very worried and rightly so'.
Yager joined OpenAI in June after a 30-year career in human rights, taking on a role focused on military applications. Her appointment followed OpenAI's contract with the Department of War in late February to deploy advanced AI systems in classified environments. That deal came after a public falling out between the DoW and Anthropic, which had sought to impose guardrails banning mass public surveillance and fully autonomous weapons. The DoW refused those terms, labelled Anthropic a supply chain risk, and OpenAI stepped in, claiming it had secured similar guardrails.
Yager admitted that joining OpenAI was not 'an immediate yes'. She was wary that the company might be hiring her as a 'branding exercise' — a figurehead without real influence. But after speaking with employees, particularly younger staff who are 'very worried about what they are creating and who's at the wheel', she felt compelled to join and see if she could make a difference. 'With apologies to my OpenAI colleagues, I am now in the belly of the beast,' she said, describing the role as a comfortable place for someone who loves to 'persuade and discuss and debate and hopefully move the needle'.
Since joining, Yager has worked directly with researchers and engineers to improve models, including ensuring appropriate responses to international agreements such as the laws of war established by the 1949 Geneva Conventions. She brought in international experts to 'literally sit at keyboards and put in prompts for really high stakes engagements', with engineers tweaking the models based on the responses. 'That had never been done before, so that was extremely gratifying,' she said.
She argued that AI companies are trying to 'reinvent the wheel' on questions of humanity and morality. 'Societies have already figured out a lot of those things, and it's called human rights. That should be the baseline,' she said. 'I do not want an AI company telling me what is good. If we allow the industry to determine what is good, then it's a free for all.'
Yager also addressed the debate in Silicon Valley about whether AI agents should have rights, a question OpenAI asked her during her interview. She said 'no' at the time and remains focused on human rights, though she is now more open to conversations about where agents fit into legal, labour and tax structures. 'This is not at all comparable to human rights, and I would really love if we could just get back to the human of human rights,' she said, drawing applause from the largely male, suited audience.
Another concern is the insular demographic building AI. She noted that last week President Trump gathered dozens of tech leaders in Washington to discuss AI safety, and all signatories of the voluntary safety accord were men, as were all attendees at a lunch event save AMD CEO Lisa Su. 'Men are awesome, but it was a very male photograph of who is going to be making the choices about AI,' she said. 'AI is fundamentally shifting the power balance between people all over the world, you did not see affected communities up there.'
Yager is also helping oversight bodies such as the Government Accountability Office and the Inspector General build tools to monitor government AI adoption, calling that accountability mechanism 'important for democracy'. Her remarks underscore the growing tension between rapid military AI adoption and the human rights principles that she argues must remain the baseline.
8
