AI Safety & Governance · Projected milestone

Automated alignment researchers

Estimated year~2029

65% by 2032 · including earlier years

The year is an editorial timing estimate. The percentage applies to occurrence by the stated deadline. These estimates are not calibrated odds.

AI systems do most alignment and interpretability research, with people setting priorities and checking results.

Explore this milestone on the interactive map

What the forecast means

Milestone stage: deployment

partially-demonstrated

Agents contribute to research, including safety-related tasks; majority verified research execution is not established.

Completion criteria: A three-month independently sampled portfolio attributes over half of completed design, implementation and analysis work to agents; publish accepted and rejected results and human repairs. This does not mean the alignment problem is solved.

This page describes a forecast, rather than an achieved result. Read the methodology and assumptions.

Built on

Circuit tracing in AI modelsRecorded · 2025enables
AI research run by agent swarmsForecast · ~2028influences
Expert-level general AIForecast · ~2028influences

Led to

Controlled recursive self-improvementForecast · ~2032enables

Sources & evidence

Research acceleration: The view inside OpenAI 2026-09-06
OpenAI reports an automated research intern, increasing agent contribution and substantial human intervention on longer tasks; it targets an automated researcher in March 2028. Internal measurements are preliminary, and compute and controls still constrain work.
Measurements for understanding the pace of AI development inside frontier labs 2026-09-17
Anthropic reports AI leading 26% of R&D in August, no fully autonomous measured R&D subset, and about 30,000 concurrent internal agents. It reports complete action-monitor coverage, which is not complete detection. The measurements use internal/model judgments.

Record reviewed 2026-10-18