As of 13 August 2026, AI can colourise your black and white photos.
Most people should do this themselves; the prompt is on this page.
Can you do it?
5 minutesto a draft.
30 minutesto something you’d act on.
Cost, all in£0
Skill neededchat-fluent
Who has to check ityou
What the alternative costsThe supplied tool data gives no price for a specialist photo editor.
If this goes wrong: the image may contain implausible colours or altered details, but you can reject the edit and try again with clearer instructions.
What to actually do
Do it yourself
The route this page recommends
A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.
How to actually do it
- Open an AI image or photo-editing tool such as Fotor and upload the highest-resolution copy of the black and white photograph.
- Gather any known information about the photograph, including its approximate date, location, people, clothing, vehicles, buildings and known original colours.
- Paste the prompt and replace the bracketed slots with the information you know, leaving unknown details marked as unknown.
- Ask the tool to create the colourised image without changing the composition or adding details.
- Compare the result with the original photograph at full size, checking faces, hands, clothing patterns, objects, text and background edges for changes or colour bleeding.
- Ask for a revised version with specific corrections, then save the version that best preserves the original and label it as colourised rather than historically verified.
Prompt
Colourise the attached black and white photograph in a natural, historically plausible way. Preserve the original faces, expressions, clothing shapes, objects, background, lighting, framing and image texture. Do not add, remove or redraw anything, and do not make the image look like a modern artificial colour filter. Use these known details where supplied: approximate date [DATE OR UNKNOWN], location [LOCATION OR UNKNOWN], known colours of clothing or objects [DETAILS OR UNKNOWN], and any other historical context [DETAILS OR UNKNOWN]. Where the colours are unknown, choose restrained realistic tones rather than bright colours. Return the colourised image and briefly list any areas where the colour is an informed guess.
Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.
Use a tool built for this
Second choiceHand it to a person
The distant thirdA person who owns the outcome does this end to end, worth it when the failure is dear.
What it gets wrong
- It cannot know the original colour of an object from greyscale information alone, so some colours are guesses.
- It can alter faces, small objects, lettering or clothing details while trying to add colour.
- It cannot recover missing photographic detail that is absent from the original scan.
- It cannot decide whether a historically unusual colour is correct without records, family knowledge or another source.
- It cannot replace your judgement about whether the result still feels faithful to the photograph.
Even on a YES, the friction has a name: taste and context depth.
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.
| Axis | Score (0–2) |
|---|---|
| Output | 2 |
| Inputs | 2 |
| Verification | 1 |
| Liability | 2 |
| Effort delta | 2 |
| Total | 9 / 10 |
The methodology and its thresholds are published in full.
FAQ
- Can ChatGPT colourise my black and white photos?
- Yes. Upload the photograph and ask it to preserve the original faces, objects, composition and detail while applying restrained, realistic colours. Treat unknown colours as informed guesses rather than facts.
- Is AI colourising old photos accurate?
- It can produce a convincing image, but it cannot know most original colours from black and white tones alone. Check the result against family knowledge, historical records or other photographs if accuracy matters.
- What is the best AI tool to colourise old photos?
- A purpose-built photo editor such as Fotor is a suitable place to start because it supports AI photo editing. A general chatbot can also help where image editing is available, but no tool can guarantee historically correct colours.
- Can AI colourise photos without changing them?
- You can instruct it to preserve the original framing, faces, objects and texture, but it may still alter small details or introduce colour bleeding. Compare the result with the original and request corrections before saving it.
Nearby answers
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.