How to Transition to AI Risk Reduction Careers | Patrick Gruban | Moral Ambition for AI скачать в хорошем качестве

How to Transition to AI Risk Reduction Careers | Patrick Gruban | Moral Ambition for AI

How can you build an impactful career in AI risk reduction? And how can you transition into AI safety from a different field? AI systems already cause real-world harms, and leading AI researchers estimate a 3–50% probability of AI causing catastrophic harm this century. Policy and technical action can reduce these risks, but too few people are working on this problem. In this Moral Ambition for AI talk, Patrick Gruban, Interim CEO of Successif, shares practical guidance on how to build or transition into a career that helps shape the future of AI for the better. He explains why AI safety needs talent from diverse backgrounds, how to assess your fit for AI risk work, how to find opportunities, and what impactful career paths look like. Successif is a free career program that pairs experienced professionals with dedicated advisors to help them transition into AI safety and governance. Register here to be informed about future talks: https://docs.google.com/forms/d/e/1FA... This talk is organised as part of the Moral Ambition for AI Circle, an initiative originating from but not organised by the School for Moral Ambition's Circle Programme. LINKS MENTIONED IN THE PRESENTATION Successif website: https://www.successif.org/ Transformer News: https://www.transformernews.ai/ AI Frontiers: https://www.ai-frontiers.org/ ML Safety: https://newsletter.mlsafety.org/ Jim Chapman’s post: https://forum.effectivealtruism.org/p... PauseAI events: https://luma.com/PauseAI SUMMARY The presentation highlights the urgent need for talent in the AI safety field, noting that leading researchers estimate a significant probability of AI causing catastrophic harm this century. While the industry currently employs only about two thousand people, it needs to recruit at least two thousand more annually to address escalating risks. The primary threats driving this demand include AI misalignment, chemical, biological, radiological, and nuclear threats, the destabilization of democracies through disinformation and surveillance, and the risks of autonomous warfare. A major misconception addressed in the talk is that entering this field requires a machine learning background or a doctorate. While mission-aligned engineers remain in high demand, the most critical bottlenecks are often in non-technical and operational sectors. Organizations desperately need research managers, directors of operations, and chiefs of staff to help scale small research teams. Furthermore, there is a strong demand for research-to-policy translators, political advocates, communications leads, and versatile amplifiers or entrepreneurial builders who can adapt to various operational and strategic needs. Gruban explains that a typical career transition into AI safety takes between six and eighteen months. He advises candidates to begin by dedicating specific weekly time to read industry newsletters and join relevant communities. Candidates should then use the Ikigai framework to identify target roles that align with their passions, strengths, financial requirements, and the world's needs. Following this reflection, individuals should upskill through foundational courses, such as those offered by BlueDot Impact, and focus on building a public portfolio of their expertise rather than relying solely on past credentials. Because half of the jobs in this sector are never publicly advertised, extensive networking and informational interviews are crucial for discovering opportunities and getting a foot in the door. To assist with this rigorous process, Successif offers free, personalized career advising specifically tailored for mid-career and senior professionals. They provide strategic guidance, resume feedback, mock interviews, and coaching to help candidates navigate rejections and overcome challenges like imposter syndrome. During the Q&A session, Gruban provided additional insights, emphasizing that while most roles are currently concentrated in the United States and the United Kingdom, professionals in other regions can leverage their local context to advocate for regional AI policies. He also advised applicants to be persistent, view the career change as a marathon rather than a sprint, and actively build their domain knowledge so they can effectively apply their existing operational or technical skills to the mission of AI risk reduction.