Notable verbatim quotes

⚠️ Quotes preserve the source's speech-to-text artifacts (garbled phrases, fragments). They are quoted exactly — do not "fix" them when re-quoting.

On "make real" and vision models (Section 2–3)

• "November 2023, GPT-4 with vision just launches. This is the first time that a really good vision model is available via like a developer API."
• "the basic idea is that you have a canvas where you can make stuff, right? You draw diagrams and things like that. And you have an AI that can see, let's see if we can get the AI to make the stuff that we're drawing."
• "For a lot of people, this was like the first time you've ever. Created something technical at all. Because it kind of code and they could use their existing skills, which was like John rectangles." (re: drawing as the input modality lowering the floor for non-coders)

On annotation as prompt input (Section 3–4)

• "This was such a cool, cool thing to discover that, that annotation could be an input to these language models."
• "if you go to some of the customers that have used the SDK now, like, like Google and mixed more stitch or prep lead and Luma, whatever Runway, this idea of using annotations as an input part of the prompt is something that has gone, people have taken really, really far."

On canvas vs chat for prompt workflows (Section 6)

• "branching conversations, as well as, like, multistep prompts. Way that it's, like, repeatable It's very hard to do in chat. Trivial to do using kind of code or some workflow editors and not exist But but hard to be creative with, I would say."
• "each one of these things is a little prompt. It has a kind of some some structure to it, but it's it's repeatable. It kinda generates its own kinda script. And then executes a script. Based on what its inputs are."

On collaboration as the canvas killer feature (Section 7)

• "the killer feature of Canvas is collaboration."
• "I wanna be able to just kinda have a 10, 15 people in the same document and for it to sort of have the affordances to to come in for all that."

On the agent harness and fairies (Section 8–9)

• "we tried to rip off cursors" (re: the 2025 agent harness)
• "each one of these is instance of that that harness that that lives of on the canvas."
• "Coordinator comes up with a to do list and then delegates those tasks. To different affairs, different agents."
• "Having a collaborative experience on the same document, together with 30 various 30 agents" (10 people × 3 fairies each)

On "code mode" and runtime API access (Section 10)

• "agents are just kinda better rather than making these calls if you just give them code and let the next it and then kinda generate the tool calls from there."
• "since it was local, we're like, let's just give the AI direct access to to the editor and let's see what it does with it."
• "the code that it uses to do it is some of the most insane using Python to template JavaScript inside of a mash call. Like, the the craziest, weirdest code that I've ever written in my life. But it totally works."
• "we've we've given it a doorway to to do that."

On monetization / experimentation posture (Section 5)

• "By the way, this is all bring your own token. Right. We haven't monetized this at all."