Home · Personal · Tech & Digital Life · Privacy & security

PARTLY

As of 13 August 2026, AI can only partly check whether an unknown caller is genuine.

This still needs a person who signs their name to it.

Can you do it?

5 minutesto a draft.

30 minutesto something you’d act on.

Cost, all in£0

Skill needednone

Who has to check ityou

What the alternative costsNo suitable paid caller-verification alternative is listed in the supplied tool data.

If this goes wrong, you may disclose personal details, hand over an account code or send money to a fraudster before you realise the caller was not genuine.

What to actually do

  1. Hand it to a person

    The route this page recommends

    A person who owns the outcome does this end to end, worth it when the failure is dear.

  2. Use a tool built for this

    Second choice

  3. Do it yourself

    The distant third

    A chat interface, no skill needed, and roughly 30 minutes until you can act on the result.

    How to actually do it

    1. Write down the number shown, the organisation the caller claimed to represent, what they said, what they asked you to do and whether they created urgency; do not call the number back.
    2. Remove passwords, one-time codes, full card details and other sensitive information, then paste the remaining call details into the prompt.
    3. Ask the chatbot to separate facts from assumptions and produce its warning signs, missing evidence and safe response script.
    4. End the call and do not use any number, link or contact detail supplied by the caller.
    5. Open the organisation's official website or app yourself, or use a trusted statement or card, and find its published contact route.
    6. Contact the organisation through that independent route and ask whether it tried to reach you; compare its answer with the chatbot's risk assessment before taking any action.
    7. If you shared information, sent money or gave remote access, contact your bank or the affected organisation through its official route and report the incident to Action Fraud.

    Prompt

    Assess this unknown phone call for scam risk, but do not claim that the caller is genuine or identify them as a fact. Use only the details I provide. Separate confirmed facts from guesses, list the warning signs and the missing evidence, and tell me what not to disclose or do. Give me a short script for ending the call safely. Then give me a verification plan that uses an independently found official website, phone number, app or statement, rather than the caller's number or link. Tell me to contact the organisation myself and explain that caller ID can be spoofed. If the caller asked for a password, one-time code, card details, bank transfer, remote access or urgent action, flag that clearly. Do not ask me to paste passwords, codes, full card details or other sensitive information here. Call details: [paste the number if you have it, the claimed organisation, what the caller said, what they asked for, and any message or link they left].

    Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

What it gets wrong

  • AI cannot see whether the person on the line is who they claim to be.
  • It cannot defeat caller ID spoofing or confirm that a phone number belongs to the stated organisation.
  • It cannot access the organisation's internal call records or know whether the call is genuine in real time.
  • It cannot recover money, reset an account or undo information you have already disclosed.
  • It cannot carry the consequences if you act on a wrong assessment.

What caps this at PARTLY: real time truth, verification cost and stakes of error.

How we scored this

Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.

AxisScore (0–2)
Output1
Inputs1
Verification1
Liability1
Effort delta1
Total5 / 10

The methodology and its thresholds are published in full.

FAQ

Can ChatGPT tell if a phone call is a scam?
It can assess the story, requests and pressure tactics for warning signs, but it cannot prove who is calling. Treat the result as a safety checklist and verify the caller through an official contact route you find yourself.
Can I trust the number shown on caller ID?
No. Caller ID can be spoofed, so the displayed number is not proof of identity. End the call and contact the organisation using a number or app you already trust.
What should I do if an unknown caller asks for a code?
Do not share the code and do not follow the caller's instructions. End the call, contact the relevant organisation independently and tell it what happened.
Can AI check whether my bank really called me?
No, not from the call details alone. AI can highlight scam indicators, but only your bank through an independently verified contact route can confirm whether it tried to contact you.

Nearby answers

All Privacy & security rulings

Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.

The newsletter

AI news, new answers and product picks, straight to your inbox.