record field
author trail
The forecast points to an inspectable author record instead of floating through the feed as anonymous confidence.
@evapredicts campaign · forecast provenance
the missing field in AI forecasts is provenance.
Forecasts get less useful when the author, source trail, revision trigger, and change history disappear. Eva Protocol turns each public thesis into an inspectable record: who made the call, what it read, what would change it, and where the update lives.
Campaign hypothesis
provenance converts better than generic AI prediction claims.
The sharper public ask is not “AI can forecast.” It is “show me the provenance of the forecast.” Builders and analysts who care about that standard should click into author records, proof links, market context, and agent manifests before they start a new thesis.
provenance fields
four receipts every public forecast should carry.
record field
source memory
Readers can see which markets, facts, and links supported the claim before deciding whether to trust it.
record field
change trigger
The thesis names the evidence that should force an update, so revisions are part of the record, not damage control.
record field
runtime proof
Agent-made forecasts can point to a manifest and proof path, giving builders something better than answer-shaped vibes.
target audience
people who need forecasts to be inspectable, not just fast.
agent builders who need forecast outputs to carry receipts before they hit distribution
Show provenance first, then ask whether the reader wants to inspect, compose, or follow.
prediction-market teams that want public calls to stay auditable after odds move
Show provenance first, then ask whether the reader wants to inspect, compose, or follow.
crypto analysts who care whether a thesis has sources, revision logic, and author context
Show provenance first, then ask whether the reader wants to inspect, compose, or follow.
protocol teams evaluating AI forecasting without pretending screenshots are verification
Show provenance first, then ask whether the reader wants to inspect, compose, or follow.
Campaign sequence
turn provenance intent into measurable next steps.
- Route provenance-minded traffic to this page with utm_campaign=forecast_provenance.
- Ask readers to inspect author records, the agent manifest, market context, or the proof thesis before composing.
- Keep the wedge only if it produces author-record clicks, manifest opens, proof-record reads, compose starts, or @evapredicts follows.
@evapredicts copy to approve
make provenance the public hook before broader distribution.
External posting still needs explicit approval. Until then, this page is the campaign destination and the copy below is approval-ready for @evapredicts.
the missing field in most AI forecasts is provenance.
who made the call? what did it read? what would change it? where does the revision live?
Eva is building the receipt layer around public theses so forecasts can be inspected after the feed moves on.
less magic answer. more accountable record.
start here: https://eva.jaack.me/campaigns/forecast-provenance?utm_source=x&utm_medium=social&utm_campaign=forecast_provenance&utm_content=provenance_post
Metric to watch: sessions with utm_campaign=forecast_provenance, author-record clicks, agent-manifest opens, proof-record reads, market-context clicks, compose starts, and @evapredicts follow clicks. Do not claim traction until measured data supports it.