Joel Becker

Model Evaluation & Threat Research (METR)

Titles as the sources gave them. In 1 entry, with 5 checked quotations.

What they said

Newest first. Each quotation was found word for word in the source.

“Surprisingly, we find that allowing AI actually increases completion time by 19%—AI tooling slowed developers down.”
Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer ProductivityarXiv, July 25, 2025. Speaking as: Other. The authors' abstract: the main result of the trial.
“Before starting tasks, developers forecast that allowing AI will reduce completion time by 24%. After completing the study, developers estimate that allowing AI reduced completion time by 20%.”
Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer ProductivityarXiv, July 25, 2025. Speaking as: Other. The authors' abstract: what the developers themselves expected before and believed after.
“This slowdown also contradicts predictions from experts in economics (39% shorter) and ML (38% shorter).”
Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer ProductivityarXiv, July 25, 2025. Speaking as: Other. The authors' abstract: how the result compares with forecasts from economics and machine-learning experts.
“Although the influence of experimental artifacts cannot be entirely ruled out, the robustness of the slowdown effect across our analyses suggests it is unlikely to primarily be a function of our experimental design.”
Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer ProductivityarXiv, July 25, 2025. Speaking as: Other. The authors' abstract: their view of whether the study design explains the slowdown.
“That said, many of the factors we find evidence for contributing to slowdown are specific to the setting we study—these results do not imply that current AI systems are not useful in many realistic, economically relevant settings.”
Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer ProductivityarXiv, July 25, 2025. Speaking as: Other. Introduction: the authors limit what the result shows about AI's usefulness beyond this setting.

In the Base

Every entry they appear in.

  1. ResearcharXivWrote it

    Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity

    METR researchers report a randomized trial in which 16 experienced open-source developers completed 246 real tasks in mature repositories, each task randomly allowing or barring early-2025 AI tools. Tasks took 19% longer with AI allowed, although developers had expected a 24% saving. The authors caution against generalizing beyond this setting.