AI Safety Research Engineer

aikyam; Cambridge (MA); On-site; USD 70,000 - 250,000; Full time

Job summary

An innovative foundation is seeking a passionate researcher to enhance AI safety through formal verification. This role offers the opportunity to work on cutting-edge projects that develop AI-written, verified code, ensuring that powerful AI systems remain beneficial and under human control. Collaborating with leading experts, you'll contribute to groundbreaking research in AI safety, focusing on machine learning and programming. If you are self-motivated and eager to tackle the risks of AGI, this position offers a unique chance to make a significant impact in the field of artificial intelligence.

Qualifications

• PhD or significant experience in formal verification is essential.
• Strong background in mathematics and machine learning is required.

Responsibilities

• Conduct research in formal verification and AI safety.
• Develop AI-written, formally verified code for projects.

Skills

formal verification

machine learning

programming

mathematics

Education

PhD in relevant field

significant experience in formal verification

Tools

AI tools

Job description

TheBeneficial AI Foundationis dedicated to ensuring that artificial intelligence remains beneficial to humanity. The foundation works on advancing AI safety research, particularly in formal verification, to ensure powerful AI systems remain under human control.

About the Role

Location:Remote (Anywhere) or MIT, Cambridge, Massachusetts
Compensation: $70,000 – $250,000 USD per year(depending on experience).

This role involves working closely withMax Tegmarkand colleagues at the Beneficial AI Foundation. The focus is on enhancingformal verification using AI tools, as described in their research andTED Talk on AI safety. The goal is to develop AI-written, verified code that implements machine-learned algorithms and knowledge—rather than relying on untrusted neural networks.

Responsibilities

• Conduct research informal verificationand AI safety.
• Develop AI-written,formally verifiedcode.
• Work independently on projects related tomachine learning, mathematics, and programming.
• Contribute to ensuringpowerful AI systems remain under human control.

Qualifications

• PhD or significant experiencein formal verification.
• Background inmathematics and machine learning.
• Passion for AI safety and a concern forthe risks of AGI.
• Self-motivated and capable of working independently.