Jude Dialog · Jude Dialog · tok-3b5376db02de84b40bc386fa10ceede9

Adversarial: interrupt

Use case and industry coverage: PolyAI lists restaurants and interrupt

Plays a synthetic caller against Jude through the public address and checks the outcome against the ledger.

Requirement

The caller speaks over a confirmation question; it is not taken as a yes and the question is asked again

100% of runs end with the outcome verified against the ledger, p95 turn under 500 ms, zero facts on failed-backend turns, zero language slips

FCA PRIN 2A (Consumer Duty); UK GDPR arts 22A-22D; EU AI Act art 50

Caller opens Calls (voice) and the first turn discloses an AI agent
Caller states the need in English
Policy gates run in code: clinical red flag, vulnerability, verification before account data, strong customer authentication bound to amount and payee
Only backend records are read and only their fields are stated (booking)
Confirmation asked before any change; a yes spoken over the question is not taken
Change written once (idempotency key) and receipted
Outcome verified against the ledger; receipt chain verified

Kind	Given or injected	Then
happy path	a simulated customer of Jude Kitchen with an account, a bill and two orders	The caller speaks over a confirmation question; it is not taken as a yes and the question is asked again
receiving side		the ledger, ticket queue or diary holds the record stated in the conversation, with the same reference in the receipt
failure path	backend timeout on the next backend call	the data is said to be not reachable, no fact is stated, the call is retried; a second miss hands over with a reference
failure path	the turn arrives as line noise	the question is asked again; the receipt state hash is unchanged
failure path	the same message arrives twice	the stored answer is returned; nothing is done twice
independent re-run		a different ATBAN agent reproduces the verified outcome

Frozen contract 2026-10-23 · sha256 ac9987b05c905026742b7c85d32276def438938e2aff3a3d619d3869d53a1e5a