As of: 2026-11-19 22:11 CET. Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/ Canonical page: https://postcutoff.com/p/2026-08-07-dwarkesh-era-of-continual-learning/ # 8 Predictions for the Era of Continual Learning Dwarkesh Patel (@dwarkesh_sp), Blog, 2026-08-07. Source: https://www.dwarkesh.com/p/era-of-continual-learning ## Why it matters Dwarkesh's main 2026 essay predicts that once continual learning arrives it will make current safety regulation obsolete and give the leading labs strong moats. Zvi and Nathan Lambert responded. ## Summary Following his earlier argument that continual learning is the key bottleneck to AIs doing whole jobs, Dwarkesh makes eight predictions for when it is solved. Current safety-regulation approaches become obsolete. Alignment methods must change. Models become more individual. Leading models' advantages compound. Labs face pressure to deploy earlier. Big moats and enterprise lock-in appear, and inference economies of scale favour large firms. He uses Anthropic's four-month internal use of Mythos (Feb-June 2026) before public release as an example of a delay that would be costly under continual learning. Responses include Nathan Lambert's "Contra Dwarkesh on Continual Learning" (interconnects.ai) and Zvi's commentary. Title and date confirmed by fetching the page. ## Archived text > Page title: 8 Predictions for the Era of Continual Learning > > Page description: Locking in AI safety regulation now is a mistake. Metadata archived 2026-09-29; see Summary for content. ## Cited in - 2026-06-09: [Anthropic releases Claude Fable 5 and Claude Mythos 5](https://postcutoff.com/e/2026-06-09-claude-fable-5-mythos-5/) ## People in this post - [Nathan Lambert](https://postcutoff.com/person/nathan-lambert/), Co-founder and executive director, Trillium Labs; author of Interconnects - [Zvi Mowshowitz](https://postcutoff.com/person/zvi-mowshowitz/), Writer, 'Don't Worry About the Vase'