Home · Personal · Family & Relationships · Relationships & social life

YES

As of 13 August 2026, AI can decide whether to end a friendship.

Most people should hand this to a purpose-built tool.

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 priced alternative for this task is listed in the supplied tool data.

If this goes wrong: you may end a valuable friendship based on a one-sided or mistaken interpretation of events.

What to actually do

  1. Use a tool built for this

    The route this page recommends

  2. Do it yourself

    Second choice

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

    How to actually do it

    1. Open a private note and write the specific events, dates or sequence of events you remember, keeping facts separate from your interpretation of them.
    2. Add what you have already said to your friend, how they responded, what boundary you want and what you would need from the friendship to continue.
    3. Paste that account into a chatbot with the prompt above, and ask it to use only the information you supplied.
    4. Answer only the follow-up questions that add information you genuinely know, and ask the chatbot to compare continuing, setting a boundary, taking distance and ending the friendship.
    5. Compare the chatbot's summary and examples against your note, deleting any claim about your friend's motives or behaviour that you did not provide.
    6. Choose the option that fits your own boundary and circumstances, then edit the proposed message so it uses your words and does not make claims you cannot support.
    7. Send the message only after deciding when and how you will handle a reply, no reply or an angry response, and speak to a trusted person first if you feel unsafe.

    Prompt

    Help me think through whether to end a friendship. Do not decide for me, diagnose either person, or assume you know the other person's motives. First summarise the situation using only the facts I provide. Then separate facts, my interpretations, repeated patterns, isolated incidents, unmet needs and possible safety concerns. Ask up to five focused questions only where the answers could change the decision. Compare these options: continue as things are, raise the issue and set a boundary, take some distance, or end the friendship. For each option, list likely benefits, costs, risks and what I would need to do next. Include a short test for whether the problem is a pattern, whether I have communicated clearly, and whether the friendship is generally mutual and respectful. Suggest a calm message for the option I choose, without blame or claims about intent. If there is intimidation, harassment, coercion or a risk of harm, prioritise immediate safety and suggest speaking to a trusted person or an appropriate support service. End with a neutral decision checklist, not a verdict. My situation is: [describe what happened, how often it happens, what I have said or tried, how the other person responded, how I feel now, and what outcome I want].

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

  3. Hand it to a person

    The distant third

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

What it gets wrong

  • AI hears one account and cannot observe the friendship, the history between you or the other person's response in real time.
  • It cannot tell whether your interpretation of an ambiguous message is fair or whether the problem is a temporary lapse or a settled pattern.
  • It cannot take responsibility for the emotional consequences of ending the friendship.
  • It cannot replace a trusted person who knows you and can notice changes in your wellbeing.
  • It can produce a balanced-looking list that gives equal weight to events that do not have equal importance.

Even on a YES, the friction has a name: judgement under ambiguity, relationship 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)
Output2
Inputs2
Verification1
Liability2
Effort delta1
Total8 / 10

The methodology and its thresholds are published in full.

FAQ

Can AI tell me whether I should end a friendship?
It can help you sort the facts, identify patterns and compare your options, but it cannot make a reliable decision for you. The decision depends on your boundaries, history and knowledge of the relationship.
Is it a good idea to ask ChatGPT about my friendship?
It can be useful for organising your thoughts or drafting a boundary-setting message. Do not treat its assumptions about your friend's motives as facts, and avoid sharing identifying or private information.
How do I know if a friendship is worth saving?
Look at whether the problem is a repeated pattern, whether you can raise it safely, and whether both people make changes after it is discussed. AI can turn those questions into a checklist, but only you can judge the history and meaning of the friendship.
How do I end a friendship without being cruel?
Keep the message brief, speak about your decision and boundaries rather than attacking the other person's character, and do not promise ongoing contact you do not want. AI can draft wording, but edit it so it sounds like you and reflects what you actually mean.

Nearby answers

All Relationships & social life 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.