OpenAI· Preparedness· San Francisco
Researcher, Recursive Self-Improvement Preparedness
Comp$295K – $445K
Classified Tasks (16)
Automate 0%Augment 69%Human-Only 31%
Augment (11)
AI assists, human decides
1. Monitor and predict the evolving capabilities of frontier AI systems.
analytical
2. Maintain and improve misuse safeguards, alignment tools, and security measures to address extreme threats.
technical
5. Analyze potential future AI risks related to recursive self-improvement and develop preparation strategies.
analytical
7. Prioritize tasks to identify work that is most useful to start immediately.
operational
8. Build rapid, scrappy prototypes and iterate on them to improve functionality and robustness.
technical
9. Execute technical research and development projects rapidly and strategically in domains with weak feedback loops.
technical
12. Track progress toward automation of technical staff by collecting and combining measures (e.g., coding time horizon, internal coding inference spending, staff surveys) to triangulate trendlines.
analytical
13. Implement data-poisoning defenses to mitigate the risk of threat actors inserting hidden objectives into models.
technical
14. Design and run experiments to assess models’ misalignment and to measure safety-relevant capability gaps.
technical
15. Develop tools for interpreting model reasoning from activations to improve model transparency.
technical
16. Prototype technical mechanisms for verifying compliance with potential future AI safety agreements.
technical
Human-Only (5)
Requires human judgment
3. Set mitigation targets by maintaining OpenAI’s preparedness framework.
leadership
4. Partner with internal staff to implement and achieve preparedness and mitigation targets.
operational
6. Translate open-ended objectives (e.g., “prepare for future security threats”) into concrete, prioritized projects and action plans.
leadership
10. Secure buy-in from other staff and stakeholders by communicating work clearly and persuasively.
communication
11. Collaborate with and manage other staff as needed to scale efforts quickly.
leadership
Job description
Researcher, Recursive Self-Improvement Preparedness | OpenAI Careers ## Researcher, Recursive Self-Improvement Preparedness Preparedness - San Francisco Apply now(opens in a new window) ## About the team Preparedness is a critical Safety Research team at OpenAI, which is focused on mitigating AI threats to global security that could scale to an extreme level of severity. Our work involves: 1. **Measurement.** Monitoring and predicting the evolving capabilities of frontier AI systems. 2. **Mitigation.** Keeping misuse safeguards, alignment tools, and security measures on track to adequately address extreme threats that might arise in the future. 3. **Coordination.** Setting mitigation targets by maintaining OpenAI’s preparedness framework, and partnering with other staff to achieve these targets. This is urgent, fast-paced work that has far-reaching implications for the company and for society. ## About the role Preparedness is hiring strong technical executors to support preparations for recursive self-improvement. This work relies on reasoning about problems that might exist in the future, but might not exist now; so it’s especially important that people in this role are tasteful and strategic. # Potential focus areas include: * **AI R&D risk measurement:** Track progress toward automation of technical staff to inform OpenAI’s near-term investments in alignment and security. For example, combine measures of coding time horizon, spending on internal coding inference, staff surveys, etc to triangulate a trendline. * **Data-poisoning mitigations:** Implement defenses that mitigate the risk of threat actors insert hidden objectives into OpenAI’s models. * **Model behavior science**: Design experiments to understand the extent to which models are problematically misaligned, or their safety-relevant capabilities lag behind dangerous capabilities. * **Model transparency:** Prepare tools for interpreting model reasoning from activations. * **Coordination and verification**: Prototype technical mechanisms for verifying compliance with potential future AI safety agreements. ## In this role, you will: * Carefully consider the problems OpenAI might face in the future and how to prepare for them. * Turn an open-ended objective like “prepare for future security threats” into a much more concrete direction (e.g. “implement monitors for data poisoning”) – prioritizing the work that is most useful to start right now. * Execute quickly, building scrappy prototypes, and then improving them iteratively. * Secure buy-in from other staff at OpenAI when necessary, and communicate your work clearly. * Collaborate with or manage other staff as needed, since we might need to rapidly scale to tackle these problems quickly. ## You might thrive in this role if you: * Are an exceptional technical executor. * Have strong strategic and research taste: you can prioritize effectively in domains with weak feedback loops. * Are passionate about mitigating the risks associated with recursive self-improvement. * Are driven by a desire to do whatever work most positively impacts the future of AI development. * Bonus: you have already done work in one of the domains listed above (ML research, AI alignment, AI verification etc). **About OpenAI** OpenAI is an AI research and deployment company dedicated to ensuring that general-purpose artificial intelligence benefits all of humanity. We push the boundaries of the capabilities of AI systems and seek to safely deploy them to the world through our products. AI is an extremely powerful tool that must be created with safety and human needs at its core, and to achieve our mission, we must encompass and value the many different perspectives, voices, and experiences that form the full spectrum of humanity. We are an equal opportunity employer, and we do not discriminate on the basis of race, religion, color, national origin, sex, sexual orientation, a