OpenAI Offers Up to $445,000 for Researcher to Tackle AI Self-Improvement Risks
मुख्य बातें
- •OpenAI is offering up to $445,000 for a researcher to model risks of AI systems that can improve themselves
- •The role requires both strong technical execution and strategic judgment to anticipate future risks
- •CEO Sam Altman has publicly set timelines for automated AI research interns by September 2024 and full AI researchers by March 2028
- •Industry leaders like Demis Hassabis and Jack Clark have recently highlighted accelerating AI capabilities and potential singularity risks
- •The Preparedness team also focuses on automated red-teaming, biosecurity, and agentic AI threats
- •Task length for frontier AI models doubles every seven months, according to METR researchers
- •The high salary reflects the premium placed on foresight and judgment in managing emerging AI risks
OpenAI has posted a rare high-paying job opening that underscores how seriously the company is treating a potentially existential question: how to build safeguards around artificial intelligence systems that may one day improve themselves without human help. The role, on the company’s Preparedness safety team, is advertised to pay between $295,000 and $445,000 annually, a salary level typically reserved for senior machine learning engineers at frontier AI labs. First reported by Business Insider, the listing seeks “strong technical executors” who are also “tasteful and strategic,” a phrase the company explains reflects the unusual nature of the work.
