The Information Machine

Productivity gains from AI are not linear, they accelerate with depth of use.

Rohan Paul Twitter · Rohan Paul (@rohanpaul_ai) · 2026-06-16

An OpenAI study finds that AI productivity gains are non-linear and accelerate as workers adopt advanced features like GPT-5 Thinking, Deep Research, and Image Generation across multiple tools and models.

Open original ↗

Appears in

Extraction

Topics: ai-productivityopenai-researchchatgptworkforce-impact

Claims

  • AI-driven productivity gains are non-linear and accelerate with the depth and breadth of AI feature usage.
  • Workers who use advanced ChatGPT capabilities — including GPT-5 Thinking, Deep Research, and Image Generation — report significantly larger time savings.
  • Multi-model and multi-tool usage correlates with greater reported productivity improvements.
  • OpenAI has published a study supporting these findings.

Key quotes

Productivity gains from AI are not linear, they accelerate with depth of use.