PRAMAN प्रमाण
On-prem · CPU-only · Air-gapped · Indic-first

Proof for what your AI says.

PRAMAN checks a generated output against the evidence it should rest on, and returns, per claim, a calibrated verdict, a distribution-free bound on the rate of auto-approving an ungrounded claim, an accept / escalate / reject decision, and a regulator-ready audit record. It bounds a rate, it is not a per-item certificate; it keeps the human on the decisions that matter.

≤ α guaranteed
Missed-approval rate
0.005/0.038/0.076
Realized FNR @ α=.01/.05/.10
100%
On-prem · no external calls
1 / claim
Tamper-evident audit record
verifier · live console
Presets
Generated output
Evidence — one passage per line
Risk level α
Policy class
Load a preset or enter an output + evidence,
then run a verification.
The method

A bound you can defend, not a vibe score.

01

Decompose

Split the output into atomic, independently checkable claims.

02

Score & calibrate

A small CPU model scores each claim against the evidence; scores are calibrated into real probabilities.

03

Conformal control

Conformal risk control picks the threshold so the missed-approval rate is provably ≤ α.

04

Decide & log

Accept the safe majority, escalate the uncertain, and write a content-hashed audit record.

Stated truthfully: PRAMAN gives provably right-sized review and a defensible audit trail. The guarantee is marginal, not a per-item certificate; it is faithfulness to the supplied evidence, not truth in the world; and it does not remove the human on catastrophic or irreversible decisions. This managed-cloud page is a demo; the product runs air-gapped on the operator's own hardware.