What is mattwood.fyi?

A live, structured feed of riffs and links from Matt Wood, Chief AI & Technology Officer at AWS — readable by people and queryable by agents as a typed knowledge graph.

Is there an API and is it free?

Yes. A public, read-only query API is available with no authentication and no API keys. See https://mattwood.fyi/openapi.json and https://mattwood.fyi/docs.

How should an AI agent use this site?

Fetch https://mattwood.fyi/feed.json for the full structured feed, or query the API for keyword/semantic search and the connection graph. See https://mattwood.fyi/agents/ for when-to-use guidance.

Thinking Fast and Slow in AI: the Role of Metacognition

AI systems have seen dramatic advancement in recent years, bringing many applications that pervade our everyday life. However, we are still mostly seeing instances of narrow AI: many of these recent developments are typically focused on a very limited set of competencies and goals, e.g., image interpretation, natural language processing, classification, prediction, and many others. Moreover, while these successes can be accredited to improved algorithms and techniques, they are also tightly linked to the availability of huge datasets and computational power. State-of-the-art AI still lacks many capabilities that would naturally be included in a notion of (human) intelligence.

arxiv.org →

Connections

Challenges

Dream-RSI: Recursive Self-Improvement through Evolving Worlds Dream-RSI's recursive self-improvement assumes agents can autonomously evolve, but this paper challenges whether narrow AI lacking metacognition can meaningfully direct its own improvement

Related to

Introducing System One Models and Jev 'System One Models' directly references the Kahneman fast/slow thinking framework that this paper applies to AI metacognition, suggesting parallel architectural thinking about dual-process AI systems
GPT-6 Astra: Looped Transformers and Hidden Reasoning GPT-6 Astra's 'hidden reasoning' and looped transformers are concrete architectural attempts at the slow deliberative thinking (System 2) that this metacognition paper theorizes AI needs

Supports

Mini-AGI: Continual Learning on 8GB VRAM Mini-AGI with continual learning represents the kind of system-level adaptation the paper argues requires metacognition to bridge narrow AI toward more general intelligence
Formal Methods for Controlling AI Agents Formal methods for controlling AI agents require the kind of metacognitive self-monitoring the paper describes, as agents need to reason about their own reasoning states and limitations
Why ML Research Agents Don't Overfit The finding that ML research agents don't overfit relates to the metacognitive regularization mechanisms this paper discusses as necessary for AI to generalize beyond narrow task performance

Type:	link
First published:	Monday, 28 September 2026 at 11:35 UTC
Location:	https://arxiv.org/pdf/2110.01834
URL:	https://mattwood.fyi/i/27e54306

I'm Matt Wood, and this is For Your Information. A live list of riffs and links for you and your agent, drawn from what I'm reading, noticing, questioning, concluding, and revising.

Connections