Otto365 AI Brief #026

Canva Craft: AI images with text you can read

How this was made: Otto365 is an animated character and his voice is generated by AI. The video tells the same story as the text on this page; the transcript lists every spoken word and the main lines shown on screen. Spotted an error?
  • Published
  • By the Otto365 AI Brief. Checked and approved by Peter Ferguson, Editor, on . The script was drafted with AI assistance, then every sentence was checked against the sources below.
  • Announced 22 Oct 2026
  • Voiced by AI

What happened

Canva's Craft promises AI images with readable text, reported by The Verge. Canva launched Craft, its own image generation model, on the 22nd of October 2026, available to all users through Dream Lab, The Verge reported.2

Three things to know

  1. Plans, then generates First, reported by TechCrunch, Canva says the model works in two stages: it reads your prompt, plans the layout, scale, visual hierarchy and text, then generates an image at up to 2K resolution in 30 aspect ratios.12
  2. Trained on templates Second, reported by TechCrunch, Canva trained it on human-designed templates, posters, social posts, presentations and menus, rather than photographs, aiming at legible text and precise edits without changing the rest of the image.1
  3. Chatbots, API planned Third, reported by TechCrunch, Canva plans to offer Craft through its API and in chatbots like ChatGPT and Gemini via its MCP, and is developing two cost-efficient versions, Craft-fast and Craft-ultrafast, no dates given.12

Why it matters to you

Otto365's take (opinion)

So, my take. AI images are turning into editable designs, and for a UK small business making menus, posters and posts, readable text is the real test: check spelling and brand details before publishing.1

Do this week

Try this: open Dream Lab in Canva, make one post in your own words, then check every word.

Sources

  1. TechCrunch: Canva releases its first image generation model (opens in a new tab) News report, accessed 23 Oct 2026
  2. The Verge: Canva’s first AI image model plans designs before generating them (opens in a new tab) News report, accessed 23 Oct 2026
The quotes we checked (20)
  • Design platform Canva released its first image generation model, called Canva Craft, developed by its research team

    TechCrunch · source 1

  • it interprets users' prompts and creates a plan with details like spatial relationships of elements, relative scale, visual hierarchy, and information for text generation.

    TechCrunch · source 1

  • its model generates an image of up to 2K resolution in a possible 30 different aspect ratios for various platforms.

    TechCrunch · source 1

  • Canva said that this approach allows it to create images with accurate details of objects and legible text.

    TechCrunch · source 1

  • users would be able to make precise edits without changing other parts of the image.

    TechCrunch · source 1

  • rather than using photographs, it used designs to train this model using human-designed templates from Canva's library, including posters, social posts, presentations, infographics, menus, and typography-rich layouts.

    TechCrunch · source 1

  • images generated with Canva Craft could be converted into layered designs using Canva Magic, the company's image-to-design transformer.

    TechCrunch · source 1

  • Canva Craft is available on the company's platform to all users, who can access the model from the Dream Lab tab.

    TechCrunch · source 1

  • The model will also be available through the Canva API. Plus, the company plans to support image generation in third-party chatbots like ChatGPT and Gemini through Canva's MCP.

    TechCrunch · source 1

  • Canva said that it is developing two cost-efficient versions of the model, called Craftfast and Craft-ultrafast, which will be released soon.

    TechCrunch · source 1

  • The company said its model works in two stages.

    TechCrunch · source 1

  • Oct 22, 2026, 9:51 PM UTC

    The Verge · source 2

  • according to an announcement on Thursday

    The Verge · source 2

  • allows it to more accurately generate complex images that require legible text and precise object placement, according to Canva.

    The Verge · source 2

  • Canva Craft is available now to all users globally.

    The Verge · source 2

  • You can try it by navigating to the three-dot option on Canva's left-hand menu and then selecting Dream Lab.

    The Verge · source 2

  • Canva Craft can generate images at up to 2K resolution and in more than 30 aspect ratios.

    The Verge · source 2

  • uses a two-stage architecture that allows it to plan a design before generating an image

    The Verge · source 2

  • The company plans to allow AI agents to plug into Canva Craft through the open Model Context Protocol, and will also roll out more efficient Craft-fast and Craft-ultrafast models in the future.

    The Verge · source 2

  • Canva is diving even deeper into AI with the launch of an image generation model that competes directly with those from OpenAI, Microsoft, and Google.

    The Verge · source 2

Transcript

Full transcript

Canva's Craft promises AI images with readable text, reported by The Verge.

Play from 0:05 Canva launched Craft, its own image generation model, on the 22nd of October 2026, available to all users through Dream Lab, The Verge reported.

Play from 0:15 First, reported by TechCrunch, Canva says the model works in two stages: it reads your prompt, plans the layout, scale, visual hierarchy and text, then generates an image at up to 2K resolution in 30 aspect ratios.

Play from 0:31 Second, reported by TechCrunch, Canva trained it on human-designed templates, posters, social posts, presentations and menus, rather than photographs, aiming at legible text and precise edits without changing the rest of the image.

Play from 0:46 Third, reported by TechCrunch, Canva plans to offer Craft through its API and in chatbots like ChatGPT and Gemini via its MCP, and is developing two cost-efficient versions, Craft-fast and Craft-ultrafast, no dates given.

Play from 1:03 So, my take. AI images are turning into editable designs, and for a UK small business making menus, posters and posts, readable text is the real test: check spelling and brand details before publishing.

Play from 1:18 Try this: open Dream Lab in Canva, make one post in your own words, then check every word.

I'm Otto365.

Follow the Brief.

On screen

Play from 0:00
Otto365 AI Brief #026, Canva Craft, 22 Oct 2026
Canva's Craft promises AI images you can read
Your prompt
Plans the design
Generates image
Legible text
Play from 0:05
Canva Craft: legible text
Launched 22 Oct 2026
Trained on: Canva templates, Not photographs
Built for: Legible text, Precise edits
Source: The Verge, 22 Oct 2026
Play from 0:15
Plans, then generates
Stage one: the plan
Spatial relationships
Relative scale
Visual hierarchy
Text generation
Stage two: the image
Up to 2K resolution
30 aspect ratios
Play from 0:31
Trained on templates
Not trained on: Photographs
Trained on: Canva templates, Posters and menus
Craft image
Canva Magic
Layered design
Edit one part
Play from 0:46
Chatbots, API planned
ChatGPT, Gemini
Canva's MCP
Canva Craft
Image generated
Craft-fast and Craft-ultrafast
No dates given
Play from 1:03
Readable text is the test
Craft makes: Legible text, Layered design
You check: Spelling, Brand details
Image to layered design
Edit one part, keep the rest
Play from 1:18
Make one post in Dream Lab
Voiced by AI. Checked by humans. Apps 365 sells AI and Microsoft 365 services. Not endorsed by companies named.

Voiced by AI. Checked by humans. Apps 365 sells AI and Microsoft 365 services. Not endorsed by companies named. Report an error.