Access to private data sources improves AI forecasting accuracy relative to public web search.
Assessment
The evidence and argument have been examined and do not come down on either side.
The claim is broad, and how true it is probably depends on the kind of question. For general event questions (geopolitics, business news, the questions asked in AI forecasting tournaments), the outcome is usually covered heavily in public reporting, so a capable agent searching the open web can likely piece together most of what a paid data feed would add. For questions about one organisation's own figures, or questions where the paid data itself is the main signal (some financial and alternative-data cases), access to that data should help, though AI agents have not been shown to make use of it.
The most direct evidence is a 2026 report credited to the CEO of FutureSearch, an AI forecasting company. It says that adding private sources such as Bloomberg terminals and proprietary datasets made no measurable difference to accuracy compared with public web search, because agents could reach the same conclusions from public secondary sources. If accurate, this weighs against the claim for general-question forecasting. But it comes from a company that sells a web-search-based product, it is not a published controlled comparison, and the original statement could not be found for this assessment. No independent benchmark that tests private data against public search for AI forecasters was found. A controlled ablation on a public benchmark, split by question type, would settle the question.
Full reasoning: the evidence and decisions behind this verdict
Prior model: the value of private data depends on how much of the signal about the question's outcome exists only behind a paywall or inside an organisation. For tournament-style questions about public events, the open web carries most of the signal, so the expected gain is small. For firm-specific or market-microstructure questions it could be large. Searches for FutureSearch, private data, Bloomberg terminals and AI forecasting found FutureSearch's material on world-models and research scaling (futuresearch.ai/findings-library/, futuresearch.ai/blog/world-modeling/) and Dan Schwarz's posts on X. None of these contained the reported "no measurable improvement from private data" statement. It is known only through the description on this claim's existing argument, so it is unverified secondhand testimony from a party with an interest. A LessWrong essay ("Bloomberg terminals for the rest of us", May 2026) is about how useful forecasting is, not about whether data sources make it more accurate. Industry pieces that assert a "proprietary data moat" (Bain, Wrike, Science Exchange) offer no forecasting-accuracy evidence. Net: there is weak evidence against the claim for general questions and no direct evidence for it, and the claim is too broad for one verdict. Hence inconclusive, with credence slightly below one half. One correction: the existing argument is labelled as supporting the claim, but its content is evidence against it.
Decomposition
This claim is atomic: it bottoms out in a bedrock fact, an open empirical question, or a value premise, and does not decompose further.
Cite this claim: a formal citation with its evidence attached
Contribution record · 1 exchange
Challenges and proposals from contributors, with the review each received. Substantive exchanges stay on the claim's public record.
The CEO of FutureSearch reported in 2026 that adding private data sources (e.g. Bloomberg terminals, proprietary datasets) produced no measurable improvement in AI forecasting accuracy over public web search, because agents reconstruct the relevant conclusions from publicly available secondary sources. Worth assessing because it bears on whether proprietary data confers durable advantage in AI-driven analysis.
The proposed canonical form, "Access to private data sources does not measurably improve AI forecasting accuracy relative to public web search," is a single reusable empirical proposition. Evidence can bear on it, and it matters for whether proprietary data gives a lasting advantage in AI-driven analysis. It concerns a public statement by a company executive, not private details about a person. The wording is neutral enough to work as a canonical form. The contributor's rationale is sincere and on topic. No sources were attached, which is not a ground for rejection at intake. Assessment will need to check the FutureSearch report against its primary source and look at other evidence on whether private data improves forecasting. The Matcher will decide whether the graph already holds this claim.
Contribute
Every judgment on this page is open to challenge. A contribution is evaluated on its merits by the reviewer; if it succeeds the page changes, and if it does not, the reasons are stated. Either way the exchange becomes part of the claim’s public record.
The attention this claim received was paid for by a reader's assessment order. Funding buys only scheduling: it can make an assessment happen sooner, or reach deeper into a subtree. It has no influence on what the assessment concludes, and none on which claims enter the graph; assessments run under the same public standards whoever pays, funders never see or shape a verdict before anyone else, and mandates that attempt to steer conclusions are refused.
Created by user · Nov 2, 2026. Every judgment on this page is accompanied by a reasoning trace.