Web Research Specialist — Frontier AI Benchmark Tasks (Turing)

Turing — the AI company working with the world's leading labs to advance frontier model capabilities — is building an evaluation benchmark for frontier AI browsing agents. Your job is investigative research: start from a verifiable fact, work backwards to construct a question that makes that fact extremely hard to locate even for a state-of-the-art AI with full web access and multiple attempts, then prove your work with a complete, auditable evidence trail.

This is not a subject-matter-expert role and not a content-writing gig. The evidence trail is the majority of the work, so it suits people with high tolerance for structured documentation: reference librarians, archival researchers, investigative journalists and fact-checkers, OSINT and due-diligence specialists, patent searchers, genealogists, and competitive quiz or puzzle-hunt constructors.

Pay	~$30 per approved task (estimate from the current task mix)
Duration	8 weeks
Commitment	Ideally up to 40 hours/week
Location	Fully remote, paid in USD
Start	As soon as you pass the assessment
Process	Apply → assessment → onboarding

Each approved task delivers a natural-language research question with a short, stable, objectively verifiable answer; independently checkable clues spanning dates, people, places, organisations, works, events, records and quantities; and a validation record of the obvious searches you ran and what they returned.