Explainers/ Research/ № 102

Verified SourceLil'Log 07-09 · 13 min read ⚑ Research paper

Members only

Full article as Markdown — paste into Obsidian / Notion

↓ Article body below · each piece has its own look

Research Explainer · XiaoHu Explains

AI Self-Improvement Starts Outside the Model: Lilian Weng on Harness Engineering

Former OpenAI safety lead surveys nearly 30 papers: from editing prompts to editing its own code, DGM pushed coding ability from 20% to 50%.

1WHY NOW

Why the Layer Outside the Model Suddenly Matters

Former OpenAI Safety Systems lead Lilian Weng published a long essay on her personal blog Lil'Log proposing "harness engineering" as a thesis, systematically connecting it to research on AI recursive self-improvement (RSI).

Primary source and reference

Lil'Log Original publisher

🔒 MEMBERS ONLY

Continue with the full member article.

That's the free preview. Behind the wall:

Full article and analysisSources and editorial notes

🔓 69 deep dives & research pieces unlock with it🎧 All podcast audio + private feed⧉ Full-text Markdown export

Unlock everything · $49/yr Monthly · $5.99 Just this piece · $1.99

Already a member? Sign in → USD checkout · Alipay / WeChat Pay for yearly & single; monthly needs a bank card · 7-day full refund on yearly.

Sign in

AI Self-Improvement Starts Outside the Model: Lilian Weng on Harness Engineering

Why the Layer Outside the Model Suddenly Matters

Primary source and reference

Continue with the full member article.

Related explainers