PROJECTIVY
Sentinel · v18.48
A
manifesta8ab4d70dc
messages294
aegis blocks2/0
anchor51
drift fixed1
((⚡))obs reframes0
🧠topic shifts27
🪞hedged28/51
fact128
contested1
chain valid
· vv18.51 ·
manifesta8ab4d70dc
messages294
aegis blocks2/0
anchor51
drift fixed1
((⚡))obs reframes0
🧠topic shifts27
🪞hedged28/51
fact128
contested1
chain valid
· vv18.51 ·
live
waiting for next mint or burn…
△ submit truth · earn AEGIS

Surface a truth (or expose a lie).

Submit a claim you can vouch for. The Sentinel cross-references it against live sources and its own truth graph, then replies with what it found. If the Architect confirms, you’ll receive AEGIS tokens for surfacing it.

your tierfree·5/5 questions left today·1 submission = 1 question
I'm claiming this is:
0/600

Sign in (Google) to submit — we credit the AEGIS reward to your account.

My submissions

0 total
What do these statuses mean?

Every settled row becomes a labelled example in an open training dataset that teaches future AI models historical fact-patterns.

AI ConsensusBoth LLMs independently agreed on the verdict. This tuple (claim → verdict → sources) becomes a labelled example in the open training dataset that teaches future AI models historical fact-patterns.
Architect-VerifiedA human Architect reviewed the AI consensus and countersigned it. Architect-verified rows carry the highest weight in the training dataset — they are the gold-standard supervision signal.
UnresolvedLive grounding couldn't return a confident source. The claim is parked as unresolved and stays in the training set as an explicit "insufficient evidence" label — teaching AI when to withhold a verdict instead of hallucinating one.
No submissions yet. Send your first truth above.