OpenAI Hiring $445,000 Researcher to Study Recursive Self-Improvement in AI

May 24, 2026

OpenAI posted a job listing in May 2026 for a safety researcher on its Preparedness team with a salary range of $295,000 to $445,000 to study recursive self-improvement in AI systems. The role supports preparations for AI models that could autonomously improve themselves. CEO Sam Altman said in October 2025 that OpenAI aims to run a true automated AI researcher by March 2028, though he acknowledged the company may fail at this goal.

OpenAI safety researcher position

▪OpenAI's safety researcher job listing seeks strong technical executors to support preparations for recursive self-improvement.
▪OpenAI's safety researcher position requires candidates to be tasteful and strategic because the work relies on reasoning about problems that might exist in the future but might not exist now.
▪OpenAI's safety researcher might make tools to interpret models' reasoning or experiment with models to understand their safety and dangers.
▪OpenAI posted a job listing in May 2026 for a safety researcher position on the Preparedness team with a salary range of $295,000 to $445,000.
▪OpenAI's safety researcher could focus on defending OpenAI's models from data poisoning attempts to corrupt an AI model through the dataset it's trained on.
▪OpenAI's safety researcher could track progress toward automation of technical staff, including measuring the usage of AI coding tools.

Recursive self-improvement capabilities

▪Google DeepMind CEO Demis Hassabis said in May 2026 that humanity now stands at the foothills of the singularity, the moment when AI begins to improve itself and outpaces human intelligence.
▪OpenAI has set the goal of making an AI tool that could research its own improvements.
▪METR researchers wrote in March 2026 that the length of a task that frontier AI models can complete doubles about every seven months.
▪Anthropic cofounder and policy head Jack Clark wrote in May 2026 that Jack Clark thinks there's a roughly 60% chance of seeing AI research and development without human involvement by the end of 2028.
▪METR wrote that AI agents will be able to handle a large fraction of the software work that takes human coders days or weeks to complete.

AI coding advancement timeline

▪Selling OpenAI's Codex coding tool to companies is a huge revenue driver for OpenAI.
▪OpenAI CEO Sam Altman said in October 2025 that OpenAI had set a goal of running a true automated AI researcher by March 2028.
▪OpenAI CEO Sam Altman said in October 2025 that OpenAI had set goals of running an automated AI research intern on hundreds of thousands of chips by September 2026.
▪OpenAI CEO Sam Altman said in October 2025 that OpenAI may totally fail at the automated AI researcher goal.
▪Anthropic published research in April 2026 on using AI models to oversee stronger AI models, with promising but limited results.

Safety preparedness measures

▪OpenAI's Preparedness team is tasked with preventing severe harms from AI.
▪OpenAI's Preparedness team job postings describe the work as urgent, fast-paced work that has far-reaching implications for the company and for society.
▪OpenAI's Preparedness team includes roles for automated red-teaming to test OpenAI's cybersecurity, biological and chemical risks, and agentic AI threats.
▪METR CEO Elizabeth Barnes wrote on May 23, 2026 that in her opinion any reasonable civilization would clearly be taking things much more slowly and carefully with AI.

Perspective of AI safety researchers and critics

▪Google DeepMind CEO Demis Hassabis said in May 2026 that humanity now stands at the foothills of the singularity, the moment when AI begins to improve itself and outpaces human intelligence.
▪METR CEO Elizabeth Barnes wrote on May 23, 2026 that in her opinion any reasonable civilization would clearly be taking things much more slowly and carefully with AI.