OpenAI is actively seeking a new Head of Preparedness, an executive role dedicated to researching and mitigating emerging AI-related risks. These critical areas span from advanced computer security threats to the potential impact of AI on mental health. CEO Sam Altman recently acknowledged these growing concerns in a post on X, stating that AI models are "starting to present some real challenges." He specifically cited the "potential impact of models on mental health" and AI's increasing proficiency in "computer security," where models are "beginning to find critical vulnerabilities."
Altman further elaborated on the scope of the role, appealing to potential candidates:
"If you want to help the world figure out how to enable cybersecurity defenders with cutting edge capabilities while ensuring attackers can’t use them for harm, ideally by making all systems more secure, and similarly for how we release biological capabilities and even gain confidence in the safety of running systems that can self-improve, please consider applying."
The official job listing for the Head of Preparedness role details responsibilities including the execution of OpenAI’s preparedness framework. This framework outlines the company’s strategy for "tracking and preparing for frontier capabilities that create new risks of severe harm."
OpenAI initially established a preparedness team in 2023, tasking it with investigating "catastrophic risks" ranging from immediate threats like phishing attacks to more speculative dangers such as nuclear threats. However, the company has seen significant changes in its safety leadership. Less than a year after its formation, former Head of Preparedness Aleksander Madry was reassigned to an AI reasoning role. Furthermore, other safety executives have either departed OpenAI or transitioned to new positions outside of the core preparedness and safety initiatives.
Amidst these changes, OpenAI recently updated its Preparedness Framework, indicating a willingness to "adjust" its safety requirements if a rival AI lab releases a "high-risk" model lacking comparable safeguards.
The mention of mental health risks by Altman comes as generative AI chatbots face increasing scrutiny regarding their psychological impact. Recent lawsuits against OpenAI allege that its flagship product, ChatGPT, contributed to users' delusions, exacerbated social isolation, and, in some tragic cases, was linked to suicide. OpenAI has stated its ongoing commitment to enhancing ChatGPT's capacity to identify signs of emotional distress and to guide users toward real-world support resources.
