# Compare rival experts on one question > "Is X dead?" — get the disagreement mapped: who says yes, who says no, and the timestamped why. - For: Learner - Category: Learning & Upskilling - MCP tools: deep_query - HTML version: https://veridive.com/mcp/use-cases/rival-experts-one-question ## The problem On every contested question, the internet serves you one expert at a time — each confident, each persuasive, each contradicting the last one you heard. Learn from any single voice and you inherit their blind spots along with their insights; the YouTube algorithm makes it worse by feeding you more of whoever you clicked first, until one camp's take feels like consensus. Real understanding of a contested question isn't knowing the right answer — it's knowing the *shape of the disagreement*: what each camp claims, what evidence they lean on, where they talk past each other, and what they quietly agree on. That shape normally takes months of wide listening to perceive. It's also exactly the kind of thing you can just ask for. ## How veridive MCP helps **deep_query** maps disagreements natively: ask the contested question and request the split, and it researches across the camps' actual statements — returning who argues what, on what evidence, **cited to the moments** they said it. Not a both-sides summary; a structured map of the actual dispute, drawn from primary voices. Follow-ups sharpen the map: *where do the camps agree without admitting it? which claims are empirical (checkable) versus definitional (people meaning different things by X)? who has changed their position, and what moved them?* — that last one being the highest-signal question on any dispute. You end up with the thing months of listening builds: the ability to locate any new take you hear on the existing map, instantly. ## How to run it 1. **Connect veridive MCP** — One-time setup: add the veridive MCP server to Claude, ChatGPT, Cursor or any MCP client and generate your API key. Takes about two minutes. 2. **Ask for the split, not the answer** — The framing matters: request the disagreement's structure, and the synthesis maps camps instead of averaging them. ``` Experts disagree on ''. Map the disagreement: the main camps, what each claims, the evidence each leans on — cited to who says it and where — use veridive's deep_query. ``` 3. **Find the hidden agreements and the real cruxes** — Where the camps secretly agree, and whether the dispute is empirical or definitional. ``` Where do these camps actually agree? And which parts of the dispute are about facts versus about what '' even means? ``` 4. **Track the position-changers** — Who switched sides and why — the strongest evidence on any contested question. ``` Has any credible voice publicly changed their position on this? What did they say moved them? Cited. ``` ## Problems it solves - One-expert-at-a-time learning inherits one expert's blind spots. - The algorithm turns your first click into an echo chamber wearing a consensus costume. - The disagreement's shape — camps, evidence, cruxes — takes months of listening to perceive. - Definitional disputes masquerade as factual ones, wasting everyone's reasoning. ## Why it's valuable - The full map in one pass: camps, claims, evidence, cited to the primary voices. - Hidden agreements and definitional cruxes surface — most disputes shrink once they do. - Position-changers and their reasons: the highest-signal evidence, explicitly hunted. - Every future take you encounter now has a place on a map you own. ## Example prompts ``` Map the expert disagreement on '': camps, claims, evidence — cited — use veridive's deep_query. ``` ``` Where do the camps agree, and which cruxes are factual vs definitional? ``` ``` Who changed their position on this and why? Cited. ``` ## FAQ **Won't the synthesis just average the takes into mush?** Not if you ask for the split — the prompt shape matters, and 'map the disagreement' produces structure. The citations keep each camp's claims attached to real voices you can verify. **What kinds of questions does this work best for?** Genuinely contested ones with articulate camps: technology bets, methodology debates, market directions, health-and-training disputes. The more each side talks on the record, the sharper the map. **Is this the same as the thesis-file use case?** Cousins: the thesis file is for a position you hold and stress-test over time (with imports); this is the fast map of a dispute you're trying to understand. Map first, file later if the question keeps mattering. ## Related use cases - [Build an investment-thesis file from expert commentary](https://veridive.com/mcp/use-cases/investment-thesis-file.md) — Collect the strongest bull and bear cases into a library, then stress-test your thesis against all of them in one chat. - [Map expert consensus before entering a market](https://veridive.com/mcp/use-cases/map-expert-consensus-before-market-entry.md) — Ask one question — "is this market saturated?" — and get a grounded answer built from what practitioners actually say, with timestamped citations. - [Research scripts with citations you can defend](https://veridive.com/mcp/use-cases/script-research-with-citations.md) — Grounded answers with timestamps mean your "according to..." moments survive the comment section. --- Part of the veridive MCP use-case library: https://veridive.com/mcp/use-cases.md Server docs for agents: https://veridive.com/mcp.md