The Idea

Most AI summaries are too fluffy. They use lots of words to say very little: "The article discusses an important development in the technology industry..." That's filler, not information.

Chain of Density fixes this with a simple constraint: write a summary, then rewrite it at exactly the same length but with more key information included. Each round, identify 1–3 important facts that are missing and work them in — which means cutting filler to make room. After about 3 rounds, you get summaries that humans actually prefer: packed with information but still readable.

Building Blocks

This composition builds on:

Extract What Matters Loop Until Done

Chain of Density combines entity identification (what key facts are missing?) with iterative refinement (keep rewriting until dense enough), constrained by fixed length to force genuine compression.

See It in Action

Summarizing a news article about Apple's iPhone 15 Pro launch.

Round 1 — Sparse
1
Start generic (~80 words)
Summary
"A major technology company has released a new smartphone that features several improvements over previous models. The device was announced at an annual event and includes upgraded internal components, a new frame material, and updated connectivity options. The phone comes with a new customizable feature."

Entities: tech company smartphone — almost no specific information.

↓ add missing key facts, same length
Round 3 — Optimal
2
Much denser (~80 words, same length)
Summary
"Apple unveiled iPhone 15 Pro at its September 2023 Apple Park event, with CEO Tim Cook highlighting the A17 Pro chip, titanium frame replacing steel, and USB-C connectivity. The $999 device features a customizable Action Button. Morgan Stanley analysts predict 80 million unit sales by year-end."

Entities: Apple iPhone 15 Pro Tim Cook A17 Pro titanium USB-C $999 80M units — packed with specifics.

Same length. Four times the information. All filler removed.

The Density Sweet Spot

Round 1
~3
Low
Round 2
~6
Medium
Round 3
~8
Best
Round 5
~13+
Too dense

Bars show entity count per summary. Humans prefer Round 3 — informative but still readable.

There's a sweet spot. Too sparse and the summary is useless filler. Too dense and it reads like a telegram. Research shows Round 3 hits the mark — humans consistently prefer it over both sparser and denser versions.

Why This Works

The fixed-length constraint is the key innovation. When you ask AI to "add more information," it naturally makes the summary longer. By holding length constant, you force it to do something much harder and more valuable: compression. To fit new facts in, it must cut filler, fuse related ideas, and choose more precise words.

This mimics what skilled human editors do: not adding more words, but making every word carry more weight. The iterative approach also prevents the model from getting overwhelmed — adding 1–3 facts at a time is manageable, even if the final density would be hard to achieve in one shot.

The Composition

Start sparse. Each round, identify missing key facts and rewrite at the same length to include them. The fixed-length constraint forces real compression — every word earns its place.

When to Use This

When to Skip This

How It Relates

Chain of Density is an iterative refinement technique, similar in spirit to Recursive Chain-of-Feedback (which iteratively fixes errors) and Reflexion (which iteratively improves through self-critique). The difference is the specific mechanism: a fixed-length constraint that forces compression rather than general improvement.

It also connects to Extract What Matters at the single-prompt level. Where that technique asks the model to identify key information once, Chain of Density does it iteratively, finding deeper and more specific entities with each round.