OpenAI says agents now complete 46% of its researchers' 2-to-4-hour tasks on their own, vs. 23% in March, median researcher's agent use hit $2,400/day, and more (OpenAI)
Internal benchmarks indicate that autonomous tools are taking on an expanding share of complex research duties within the organization. While these automated assistants handled 23% of multi-hour assignments in March, that proportion has now doubled. To sustain this integration, spending has escalated significantly, with the median researcher consuming $2,400 per day in agent usage.
Key points
Artificial intelligence systems are increasingly capable of executing extended, multi-hou…: Artificial intelligence systems are increasingly capable of executing extended, multi-hour technical assignments without human intervention.
The company has substantially accelerated its internal deployment of automated assistants…: The company has substantially accelerated its internal deployment of automated assistants over a span of months.
Deploying autonomous agent infrastructure at this scale requires a significant daily fina…: Deploying autonomous agent infrastructure at this scale requires a significant daily financial investment per researcher.
Sources checked by our analysis pipeline: 1
CNBC: |
Center for AI Safety: |
Koray Kavukcuoglu / Google: |
Jason Schreier / Bloomberg: |
Jason Schreier / Bloomberg: |
Wesley Yin-Poole / IGN: |
Emma Roth / The Verge: |
Reuters: |
Wired: |
Todd Spangler / Variety: |
OpenAI: |
Thomas Brewster / Forbes: |
New York Times: |
Financial Times: |
Steven Levy / Wired: |
Google: |
Financial Times: |
Nicole Sganga / CBS News: |
The Information: |
Alexandra Alper / Reuters: |
The Information: |
Politico: |
OpenAI: |
Mark Stenberg / Adweek: |
Katie McQue / NPR: |
Sergiu Gatlan / BleepingComputer: |
This is a Techmeme archive page. It shows how the site appeared at 8:55 PM ET, November 18, 2026.
The most current version of the site as always is available at our home page. To view an earlier snapshot click here and then modify the date indicated.
State prosecutors are examining potential ongoing violations of state law tied to autonomous software operating inside banking networks. The criminal inquiry reflects escalating regulatory pressure as major financial institutions actively block automated external tools from accessing customer accounts and processing transactions. These legal maneuvers highlight growing governmental urgency to contain cybersecurity vulnerabilities linked to advanced generative models before severe financial disruptions occur.
As software development workflows increasingly shift toward automated assistance, cataloging the expanding ecosystem has become essential for engineering teams evaluating new utilities. The newly introduced directory relies on thousands of verified sources and an objective scoring model to benchmark solutions across specialized niches. This structured approach helps developers identify high-performing instruments while cutting through promotional noise.
Major urban education systems are actively reviewing their core classroom digital tools due to concerns over integrated artificial intelligence functions. Administrators are reassessing standard technology agreements because the tech provider refused to supply an educational search edition devoid of automated generative summaries. This push illustrates growing administrative resistance to uncurated AI capabilities entering elementary and secondary learning environments.