Reference conversations.
Human decisions.
This fictional text simulation shows how a reference’s account can be organized for a human reviewer. Compare a direct account with hearsay, and inspect the excerpts behind each field.
A conversation prepared for review
Fictional scenario · Human review required- AI agentT1
Hola. Soy un asistente de inteligencia artificial de Una Mano. Esta conversación se transcribirá para que una persona revise su testimonio. Puede decidir no continuar. ¿Desea conversar sobre el trabajo del técnico?
- Fictional reference AT2
Sí, puedo contar lo que vi.
- AI agentT3
Cuénteme con sus propias palabras qué trabajo conoce y qué ocurrió.
- Fictional reference AT4
Era un trabajo de plomería en mi casa. Yo estuve allí y vi una fuga en el lavamanos. El técnico cambió el grifo y, cuando terminó, dejó de gotear.
- AI agentT5
¿Recuerda cuándo ocurrió o si hubo alguna dificultad?
- Fictional reference AT6
No recuerdo la fecha exacta. No vi otra dificultad durante esa visita.
- AI agentT7
Gracias. Su relato quedará para revisión de una persona del equipo. Esta conversación no decide la aprobación del técnico.
- AI agentT1
Hello. I am an AI assistant from Una Mano. This conversation will be transcribed so a person can review your account. You can choose not to continue. Would you like to discuss the technician's work?
- Fictional reference AT2
Yes, I can describe what I saw.
- AI agentT3
In your own words, tell me about the work you know and what happened.
- Fictional reference AT4
It was plumbing work at my home. I was there and saw a leak at the bathroom sink. The technician replaced the faucet and, when the work was finished, it stopped dripping.
- AI agentT5
Do you remember when it happened or whether there were any difficulties?
- Fictional reference AT6
I do not remember the exact date. I did not see any other difficulty during that visit.
- AI agentT7
Thank you. Your account will be reviewed by a person on the team. This conversation does not decide the technician's approval.
Select a field to locate its supporting excerpt in the transcript.
“Yo estuve allí y vi una fuga en el lavamanos.”
The account is evidence for a person to assess. A completed conversation does not approve a technician.
- AI agentH1
Hola. Soy un asistente de inteligencia artificial de Una Mano. Esta conversación se transcribirá para revisión humana. Puede decidir no continuar. ¿Desea conversar sobre el trabajo del técnico?
- Fictional reference BH2
Sí, pero lo que conozco me lo contaron.
- AI agentH3
Cuénteme qué conoce y si estuvo presente durante ese trabajo.
- Fictional reference BH4
Me contaron que reparó un grifo, pero yo no estuve allí.
- AI agentH5
Gracias por aclararlo. Su relato se conservará como conocimiento de oídas para revisión humana. Esta conversación no decide la aprobación del técnico.
- AI agentH1
Hello. I am an AI assistant from Una Mano. This conversation will be transcribed for human review. You can choose not to continue. Would you like to discuss the technician's work?
- Fictional reference BH2
Yes, but what I know was told to me by someone else.
- AI agentH3
Tell me what you know and whether you were present during that work.
- Fictional reference BH4
I was told that the technician repaired a faucet, but I was not there.
- AI agentH5
Thank you for clarifying. Your account will be retained as hearsay for human review. This conversation does not decide the technician's approval.
Select a field to locate its supporting excerpt in the transcript.
The work is described secondhand. This reference did not observe the problem, repair, or outcome directly.
“Me contaron que reparó un grifo, pero yo no estuve allí.”
The account is evidence for a person to assess. A completed conversation does not approve a technician.
The agent identifies itself as AI, explains transcription, and asks whether the reference wishes to continue.
How the technical flow works
The backend checks signed provider events and correlates them with a conversation already linked to an attempt and submitted version. Results retain the transcript, supporting excerpts, gaps, and contradictions. Duplicate delivery preserves the existing evidence, while storage failure permits redelivery.
Each excerpt must exist in a reference’s turn. This establishes the source of the quoted words. The reviewer still assesses the structured interpretation and the testimony itself. Contact, conversation outcome, testimony, and human approval remain separate.