METR Research Shows Frontier AI Task-Completion Capabilities Doubling Every Few Months
Worth a glance, not a chew.
Summary
This article from LessWrong discusses METR's findings that the length of tasks frontier AI models can reliably complete has been doubling every few months. It explores the concept of "No-CoT" (no chain-of-thought) task-completion time horizons for advanced AI models, analyzing how rapidly AI capabilities are expanding in terms of autonomous task execution without explicit reasoning chains.
Key quotes
· 1 pulledAbout a year ago, METR showed that the length of tasks frontier models can reliably complete doubles every few mo…
You might also wanna read
AI Task Completion Capabilities Show Exponential Growth, Could Handle Most Software Tasks Within a Decade
The article presents a methodology for measuring AI performance based on the length of tasks AI agents can complete independently. It shows
metr.org·5mo ago
METR Adjusts Developer Productivity Experiment Design Due to Unreliable AI Impact Data
METR is changing its developer productivity experiment design after finding that new data from August 2025 gives an unreliable signal about
metr.org·3mo ago
METR Study Finds AI Coding Tools May Not Boost Developer Productivity as Much as Claimed
A METR experiment conducted in spring 2025 challenges the prevailing narrative that AI coding tools dramatically boost developer productivit
secondthoughts.ai·11mo ago
The Coming Scarcity of Frontier AI Access
The article argues that the common belief in widespread, abundant access to frontier AI models is misguided. Contrary to the mantra that AI
writing.antonleicht.me·1mo ago
AI Evolution in 2025: From Stochastic Parrots to Chain of Thought Reasoning
The article reflects on the evolution of AI understanding by the end of 2025, noting that the 'stochastic parrots' criticism of LLMs has lar
antirez.com·5mo ago
Examining the Limitations of Transformer Models and the Gap to Human-Level AI
The article presents a skeptical perspective on claims about imminent Artificial General Intelligence (AGI), arguing that current transforme
dlants.me·4mo ago