TL;DR
An LLM compresses all human knowledge into a single distribution — and with every new request it gravitates toward the center. Ask a model ten times to come up with business ideas and you'll get ten similar answers clustered around "typical" solutions. This isn't a bug in a particular model — it's a fundamental property of how LLMs are trained. Two mechanisms work against diversity: fixation within a single session and knowledge aggregation across sessions.
Fixation is when early ideas suck in the later ones. The model generates text step by step: each next token depends on the previous ones. If the first idea is "a gamified app," the next four will most likely circle around gamification too. On top of that, RLHF (reinforcement learning from human feedback) pushes the model further toward "familiar" answers: annotators more often approved the predictable, and the model absorbed that.
Two targeted prompts treat each mechanism separately: Chain-of-Thought (CoT) breaks fixation within a session — the model first sketches titles, then forcibly diversifies them, then expands them. Ordinary personas (not "think like Steve Jobs," but "you're an elementary school teacher who runs marathons") pull the model into different corners of its knowledge across sessions. Together they deliver diversity above the human level.
Want this for your own case? Apply this study to your task: AI adapts its method to what you are working on and writes the prompt.
Method Blueprint
Two levels that work independently — and together give the maximum:
LEVEL 1: CoT — against fixation (one request)
STEP 1: Generate {N} short idea titles (don't expand them)
STEP 2: Revise the titles — make each one as different
from the others as possible, replace repeated themes
STEP 3: Expand each title into a full idea
LEVEL 2: Personas — against aggregation (several requests)
REQUEST 1: [Persona A — specific, ordinary] + the CoT above → ideas
REQUEST 2: [Persona B — different, ordinary] + the CoT above → ideas
REQUEST 3: [Persona C — different, ordinary] + the CoT above → ideas
...
OUTCOME: merge all the idea sets → maximum coverage
⚠️ Each level works in a regular chat. CoT is a single request.
Personas are separate requests (or one with several roles in a row).
Example in Practice
Task: Come up with ideas for Substack newsletters in the personal finance niche. Without the technique, the model churns out the same thing every time: "investing tips," "stock breakdowns," "financial literacy." Boring and predictable.
Prompt (CoT + persona):
You are Megan, 34, a kindergarten teacher from Toledo, Ohio.
You love vegetable gardening and Nordic walking. You got into personal finance
two years ago after your divorce — you learned everything from scratch.
Your task: come up with ideas for personal finance newsletters on Substack.
STEP 1. First, sketch 8 short concept titles (3–7 words each).
Don't expand them — titles only.
STEP 2. Reread the list. Find repeated themes and angles.
Replace similar titles with fundamentally different ones —
so that the 8 ideas cover the widest possible range of the topic.
Show the updated list.
STEP 3. Expand each idea: 2–3 sentences — what the newsletter is about,
who the audience is, how it stands out from the rest.
Then repeat the request with a different persona — for example, "Mike, 52, an Army veteran from Tucson who retired early" or "Katie, 26, a freelance designer from Chicago who works for overseas clients."
Result:
The model will show all three stages explicitly: a rough list of titles → a revised list explaining what it changed and why → fully developed concepts. Each persona pulls toward its own corner: the teacher gives ideas through the lens of "an ordinary person with no finance background," the veteran through discipline and early retirement, the freelancer through irregular income and currency risk. Merge the three sets and you get a palette no single request will ever give you.
Why It Works
The fixation problem. An LLM generates text step by step — each token depends on the previous ones. That means the first idea literally "weighs" more when generating the second, the second when generating the third. Start with "gamification" and the next ideas will gravitate toward that theme. Plus, RLHF taught the model to produce "approvable" answers — and approvable means familiar and predictable.
Why CoT breaks fixation. Step 1 (titles only) keeps early ideas from "growing" and taking over the context. Fewer tokens → a weaker anchor. Step 2 (an explicit instruction to diversify) is a direct command to break the "typical" pattern. LLMs follow instructions well — unlike people, for whom being told "come up with something different" barely helps.
Why "Steve Jobs" is worse than "kindergarten teacher." "Steve Jobs" and "creative entrepreneur" are central, recognizable concepts with a very specific aura: innovation, design, "think different." The model knows that image well — and gravitates toward its center. An ordinary persona with a specific biography ("a divorced kindergarten teacher from Toledo") is a particular corner of the knowledge space that the model uses as an anchor in an atypical spot. The more specific and ordinary, the farther from the center.
Control levers: - Number of CoT steps → can be cut to 2 (titles + straight to expansion) for simple tasks - Persona specificity → the more biographical detail, the stronger the anchor; a faceless role ("a student") barely helps - Number of personas → more personas = wider coverage; 3–5 different ones give a noticeable jump - Explicitness of the step 2 instruction → you can add "list which themes are already used and what's missing" — the model will make the diversity visible
Prompt Template
You are {name}, {age}, a {profession} from {city}.
{1-2 specific details: hobbies, life experience, context}.
Task: come up with ideas for {task topic}.
STEP 1. Sketch {N} short concept titles (3–7 words).
Titles only — no explanations.
STEP 2. Reread the list. Find repeated themes and angles.
Replace similar ones with fundamentally different ones — so the {N} ideas
cover the widest possible range.
Show the updated list.
STEP 3. Expand each idea: {format — e.g., 2-3 sentences /
bullets / a table}.
What to fill in:
- {name}, {age}, {profession}, {city} — specific and ordinary; not "Elon Musk," but "Tom, an accountant from Dayton"
- {1-2 details} — hobbies, life experience, something specific
- {N} — 6–10 for good coverage
- {task topic} — as specific as possible: "ideas for newsletter posts about remote work," not just "content ideas"
- {format} — whatever output format is convenient for you
🚀 Quick start — paste into the chat:
Here's a template for generating diverse ideas with CoT + a persona.
Adapt it to my task: {your task}.
Ask me questions to fill in all the fields.
[paste the template above]
The LLM will ask about the task topic, the desired format and the number of ideas — because without them it doesn't know what kind of persona to create or how many steps to run.
Limitations
⚠️ A method for ideation, not facts: CoT + personas help where you need broad coverage of options. For factual questions ("how much does it cost to register an LLC?") the technique is pointless.
⚠️ A persona anchors but doesn't guarantee: if the personas are too similar ("a manager from Chicago" and "a marketer from Chicago"), diversity will be minimal. You need contrasting biographies — different ages, professions, life contexts.
⚠️ Raising the temperature doesn't help: it's tempting to crank up randomness in the settings. But according to the study, a small temperature increase gives a minimal gain in diversity, and a high temperature wrecks quality: the model starts producing nonsense.
⚠️ CoT doesn't work for people the way it does for LLMs: if you're brainstorming with a team, this technique won't help humans equally. People are worse at following a "make it different" instruction — and still gravitate toward what they've already come up with.
⚠️ The maximum effect takes several requests: a single CoT request improves diversity within a session. The real breakthrough is several requests with different personas. It takes extra time, but the result is worth it.
How They Studied It
The Columbia Business School team ran a "fair" experiment — unlike most earlier work, where LLMs and humans got different instructions. Here the conditions were equalized: the same tasks, the same process structure. Four studies, thousands of ideas.
The main methodological contribution is a new way to measure idea diversity. Standard methods compare word similarity (embeddings). Instead, the researchers used an LLM for semantic categorization: the model sorted ideas into conceptual categories rather than by lexical similarity. That's more accurate: "a meditation app" and "a mindfulness program" are lexically different but semantically the same thing.
To measure fixation they looked at the rate at which new categories accumulate: if the first three ideas come from new categories and the next seven repeat old ones, there's fixation. The LLM showed the same curve as humans: diversity piles up quickly at first and then fades. CoT flattened that curve — the model kept adding new categories for longer.
The most unexpected result: "Steve Jobs" is worse than an ordinary persona. Intuition says the opposite — a creative icon should produce unconventional ideas. But it turned out this image is too "central": everyone knows what Jobs-style thinking is, the model knows the pattern well and reproduces it predictably. An ordinary persona with a specific biography is a rarer angle that the model illuminates differently.
The CoT + ordinary personas combination exceeded the level of human diversity for the first time in the literature. It's not just an "improvement" — it's a change in logic: LLMs used to be catching up with people, and now, with the right prompts, they pull ahead.
Original from the Paper (optional)
The authors describe the CoT method they tested as follows (based on Meincke et al. 2024b):
Chain-of-Thought prompting for ideation:
Step 1: Generate a set of short idea titles (summaries),
without expanding them into full ideas.
Step 2: Explicitly revise the titles to ensure they are
different from each other — bolder, more distinct,
covering different angles.
Step 3: Expand the revised titles into full ideas.
Key mechanism:
- Short titles prevent early elaboration → weaker token anchoring
- Explicit "make different" instruction → breaks typicality bias from RLHF
Context: This is an operationalization of CoT for ideation tasks drawn from several earlier works, which the authors tested against other methods.
Adaptations and Extrapolations
💡 Adaptation: Several Personas in One Request
If you don't want to run three separate requests, ask the model to play several roles in sequence within one conversation:
We're going to run three rounds of ideation for {task}.
In each round you take on a new role and generate ideas using the 3 steps below.
ROUND 1: You are {Persona A}. [CoT steps 1-3]
ROUND 2: You are {Persona B}. [CoT steps 1-3]
ROUND 3: You are {Persona C}. [CoT steps 1-3]
After three rounds: pick the 5 most dissimilar ideas across all rounds.
The loss is small (one shared context instead of three clean ones), but you win on convenience — everything in a single request.
🔧 Technique: Add an "Antagonist" to CoT
In step 2 you can add an explicit list of what's already been covered:
STEP 2. List all the themes and angles from step 1.
Now find 3 themes that are NOT on this list.
Replace 3 similar titles with ideas from these new themes.
This makes the diversification process visible and mandatory — the model literally maps what it has already used and hunts for blank spots.
🔧 Technique: The Outsider Persona
The most powerful persona is someone from a different industry looking at your task for the first time:
You are Linda, 47, an agronomist from Fresno, California.
You just found out what Substack newsletters are, and you're thinking about
how to apply what you know about farming
to the topic of {task}.
The farther the outsider is from the topic, the more unexpected the analogies. An agronomist in the personal finance niche will offer ideas about "crop rotation" for capital, diversification as polyculture, the "yield" of investments — metaphors a "finance expert" would never reach.
Resources
Examining and Addressing Barriers to Diversity in LLM-Generated Ideas Yuting Deng, Melanie Brucks, Olivier Toubia Columbia Business School
The work builds on: - Meincke et al. 2024b — CoT for ideation - De Freitas et al. 2025 — a review of prompting strategies for creativity - Doshi & Hauser 2024 — the "tragedy of the commons" in the market for ideas - Wei et al. 2022 — Chain-of-Thought prompting (the original)