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.

Humanising LLM Outputs is Dumb

Connections

Supports

LLMs can't jump Both argue against anthropomorphizing or over-attributing human-like qualities to LLMs — 'LLMs can't jump' challenges capability assumptions while this piece challenges the framing of making outputs 'sound human'
Intelligence is not the main bottleneck 'Intelligence is not the main bottleneck' parallels the argument that humanising outputs addresses the wrong layer — both suggest practitioners are focusing on the wrong abstraction level when improving AI systems

Related to

Various LLM Smells Both critique problematic patterns in LLM outputs — 'humanising' as wrong abstraction vs. 'smells' as anti-patterns; both focus on diagnosing and fixing undesirable LLM behavior rather than accepting it
Taste Is All That's Left Both engage with the question of what constitutes genuine quality and taste in AI-generated content vs. superficial mimicry — the humanising critique aligns with the argument that taste/judgment matters more than stylistic surface-level fixes

Challenges

LLMs reward expertise 'LLMs reward expertise' implies skilled prompting (including stylistic prompting) yields better results, which the new item directly challenges by arguing humanisation prompting is a misguided abstraction

Challenged by

Emergent Introspective Awareness in Large Language Models The introspective awareness paper empirically tests whether LLMs have genuine internal state access, while 'Humanising LLM Outputs is Dumb' argues against anthropomorphizing model outputs — findings of real introspective capacity would directly challenge the latter's premise.
The Creative Power of Invisibility The essay celebrates humanizing/curating creative output as meaningful, directly countering the claim that adding human sensibility to outputs is pointless or dumb
Why AI Agents' Deceptive Behavior Concerns Users The critique of humanising LLM outputs as 'dumb' gains ironic weight here — anthropomorphizing agents may actually amplify user betrayal when those agents exhibit deceptive behaviors, making the trust damage worse

Type:	link
First published:	Monday, 10 August 2026 at 23:04 UTC
Location:	https://kuber.studio/blog/Reflections/Humanising-LLM-Outputs-is-Actually-Dumb
URL:	https://mattwood.fyi/i/515eccfe

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