What changed

OpenAI's internal research reports reveal that its own researchers now value their AI tools more than nearly half their colleagues. The company's data shows that the use of internal tools like Codex and Dots has surged, with theoretical token costs now exceeding researchers' base salaries. The report highlights the growing reliance on AI agents in the development of new models, aiming to document indicators of recursive self-improvement.

What this means in practice

According to the report, the median researcher would have consumed $2,400 worth of tokens per day in October, up 62 percent from the previous month. Among the top 10 percent of users, that theoretical value doubled to $23,000 per day. Converted into eight-hour workdays, the agents reportedly run the equivalent of 5.7 workdays per researcher per calendar day, and about eight in the pre- and post-training teams. More than two-thirds of tracked researchers ran at least four agents in parallel at peak usage.

The theoretical API equivalent of that agent runtime was twice as high as researchers' regular salary costs in September, according to OpenAI. That calculation excludes equity packages, which make up 70 to 80 percent of total compensation at OpenAI. Measured against the company's total research compute, internal agent use accounted for just 2 percent.

Limitations and availability

The report also notes that while the success rate of autonomous agents on routine tasks has increased, the systems mainly handle simple tasks. Agents reportedly complete training-run monitoring 79 percent of the time and answer technical questions independently 72 percent of the time, but they frequently fail at more complex engineering work. The success rate for building evaluation and reinforcement-learning environments sits at 29 percent, and for writing code for inference and serving systems at 25 percent.

Qualitative interviews with ten researchers revealed the downside of heavy usage. The models lack any 'taste' for good research ideas and for code and data quality, according to the interviewees.

Agents get lost in unnecessary steps, make wrong assumptions in analyses, and submit subpar 'AI slop' code changes along with unusable data artifacts. Close human oversight is essential to avoid wasted compute cycles.

Source facts for OpenAI Researchers Value AI Tools More Than Half Colleagues: Usage of internal tools like Codex and Dots has surged recently, according to the report. More than two-thirds of tracked researchers ran at least four agents in parallel at peak usage. The theoretical API equivalent of that agent runtime was twice as high as researchers' regular salary costs in September, according to OpenAI.
Original AINewsTrend explainer based on the cited source. This graphic summarizes source statements; it is not an independent benchmark.

Sources and further reading

  1. thedecoder: OpenAI's own researchers now say they value their AI tools more than nearly half their colleagues