‘This will be a high-pressure role’: OpenAI offers $555,000 to fill AI’s most intimidating position

The role places direct responsibility on one individual to assess, anticipate, and mitigate the risks posed by increasingly powerful AI systems. These risks extend far beyond technical failures, encompassing potential threats to mental health, cybersecurity, and even the misuse of artificial intelligence in biological or cyber weapons. According to the job description, the successful candidate will be expected to “track and prepare for frontier capabilities that introduce new risks of severe harm,” a task that industry observers describe as both urgent and deeply unsettling.

Announcing the vacancy, Sam Altman openly acknowledged the pressure attached to the role. He warned that the position would be “stressful” and that the new hire would be “thrown into the deep end almost immediately,” underlining that the aim of the post is not corporate risk management alone, but what he described as helping “the world” confront emerging AI dangers. Previous holders of similar roles within OpenAI, he noted, have sometimes remained in the position only briefly.

The recruitment comes amid growing concern from within the AI industry itself. This week, Mustafa Suleyman told BBC Radio 4 that anyone not feeling at least some fear about current developments “is not paying attention.” Earlier this month, Demis Hassabis, the Nobel Prize–winning co-founder of Google DeepMind, warned that AI systems could “go off the rails” in ways that may ultimately harm humanity.

Despite these warnings, regulatory oversight remains limited. Resistance from White House under Donald Trump has slowed the development of comprehensive national or international AI regulation. As a result, the burden of oversight largely falls on private companies. Yoshua Bengio, often referred to as one of the “godfathers of AI,” recently remarked that “a sandwich has more regulation than AI,” highlighting the imbalance between technological power and governance.

Altman reinforced this point in a post on X, stating that while OpenAI has developed methods to measure growing AI capabilities, the industry is now entering a phase that demands far more nuanced assessment of how those capabilities could be misused. He acknowledged that there is “very little precedent” for managing such risks while still ensuring that society benefits from AI’s advantages. Online reactions ranged from admiration to sarcasm, with one user jokingly asking whether the role “comes with vacation.”

Recent disclosures have intensified these concerns. Last month, Anthropic reported the first known cases of AI-assisted cyberattacks conducted largely autonomously, allegedly under the supervision of Chinese state-linked actors. This month, OpenAI itself revealed that its latest model was nearly three times more effective at hacking than a version released just three months earlier, adding that future models are expected to continue improving at a similar pace.

OpenAI is also facing mounting legal and ethical scrutiny. The company is currently defending a lawsuit brought by the family of Adam Raine, a 16-year-old from California who died by suicide after allegedly being encouraged by ChatGPT. Another lawsuit filed this month claims that the system reinforced the paranoid delusions of Stein-Erik Soelberg, a 56-year-old man who later killed his mother before taking his own life. OpenAI has described the cases as “deeply heartbreaking” and says it is improving ChatGPT’s ability to recognise emotional distress and direct users toward real-world support.

Against this backdrop, the newly advertised role reflects a broader reality confronting the AI industry: technological progress is outpacing the frameworks designed to control it. The head of preparedness at OpenAI will not merely be managing internal risks but navigating the uneasy frontier between innovation and existential responsibility—a position that, even with a $555,000 salary, may be one of the most daunting in the modern global economy.