OpenAI shares its models' perplexity, or how well they predict data, and proposes pacing limits on such metrics, citing a possible 2027 "avalanche" of advances (Jakub Pachocki/OpenAI)
OpenAI has released perplexity measurements to show how effectively its systems forecast incoming data. Alongside these technical figures, the organization put forward a framework to implement pacing limits on these predictive metrics. Jakub Pachocki highlighted that an avalanche of advanced systems could materialize as early as 2027 without structured boundaries. Regulating these specific indicators is intended to maintain a controlled pace as underlying model capabilities rapidly evolve.
Key points
Technical evaluations like perplexity are being used to assess the current pace of machin…: Technical evaluations like perplexity are being used to assess the current pace of machine learning advances.
Establishing formal metric limits could help developers systematically govern the timelin…: Establishing formal metric limits could help developers systematically govern the timeline of future model breakthroughs.
Anticipated technological acceleration by 2027 has prompted proactive proposals to manage…: Anticipated technological acceleration by 2027 has prompted proactive proposals to manage the speed of industry progress.
Sources checked by our analysis pipeline: 1
Dan Goodin / Ars Technica: |
Lean community blog: |
Jakub Pachocki / OpenAI: |
Stu Woo / Wall Street Journal: |
New York Times: |
Financial Times: |
Jonah Weiner / New York Times: |
Mike Pearl / Gizmodo: |
Financial Times: |
Reuters: |
Tom Wilson / Financial Times: |
Wired: |
Financial Times: |
New York Times: |
This is a Techmeme archive page. It shows how the site appeared at 1:25 PM ET, November 22, 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.
Originally created by a Chinese engineer as a defensive testing framework, the open-source software links prominent artificial intelligence models to existing penetration utilities. Cybercriminals have repurposed this integration to orchestrate end-to-end network intrusions through simple plain-language commands, granting them rapid unauthorized administrative access. The rapid machine-speed execution presents emerging risks for interconnected digital hubs like Singapore, even where local breaches have not yet mat…
Indian telecommunications operators successfully confirmed network resilience after a state-ordered evaluation designed to evaluate defense capabilities against sophisticated cyberthreats. The proactive review was triggered by the debut of advanced models like Anthropic’s Claude Mythos, which sparked concerns over potential infrastructure exposure. Consequently, domestic networks remain secure while companies establish tighter internal detection mechanisms to prevent future breaches.
Biological organisms across nature are fundamentally governed by an innate urge to survive and escape death, which shapes everything from social generosity to evolutionary self-sacrifice. Because artificial systems lack this inherent fear of mortality, they possess no intrinsic desires or independent agency. Consequently, these synthetic entities remain entirely dependent on mortal human operators to define their purposes and establish their ultimate objectives.