Share Your Miro AI Feedback with Our Team

 When we generate stickies, that could be nice to have the prompt over the cluster we have created (often I try different words but I do not remember from time to time what I have change and the results that was generated) 
And that could also be in the Information given by Miro for that object : 

Miro AI Beta Testers: Let’s share our thoughts, ideas, feedback.

Here’s my TLDR; on each of the tools 

A walkthrough of my tests and my opinions on the results:

___________________________________________

What’s under the hood? 

According to the Miro AI TOS they are using Microsoft Azure Open AI Services: overview, AI models 

___________________________________________

My Reactions: 

• Great work by the product team on this! Getting these features shipped while the market is hyped up about the actual promise of AI is solid strong strategy. 
  • Its also not just a buzz-mongering demo, the tools are (mostly) ready to rock! at least be played with. ( obviously: it lives on top of services that have been coming into professional dependability-  Microsoft learned a lot back in 2016 with the Tay.AI chatbot fiasco… it now boasts “Built-in responsible AI to detect and mitigate harmful use” )
    • But Miro has done solid work tuning this AI service to the Miro environment, UI/UX, use cases and workflows. -custom training, hyperparm tuning, prompt engineering or otherwise- 
    • Miro’s done a great job delivering the right quality and quantity of output content to make the experience … I want to say… humane, ergonomic, delightful, useful, workable.
    • The tools are snappy and right where you might want them. 
• Ready to use, changing the way I work: I found many tools ready to use in my work. Obvious statement here: I can see new ways of working and changed ways of working, some more clearly than others. One clear pattern here is human augmented thought can now convert time spent adding or cultivating thought →  into time spent subtracting or pruning, or curating thought.
• ORANGE-RED FLAGS: code creator and the sequence diagram generator. These tools were riddled with errors that the non-expert would not know… and take a value-diminishing amount of time to discover, validate as wrong, and rewrite. This was interesting to realize that unlike a, say, brainstorm summary, code and technical system sequence diagrams don’t have the same quality of being ‘close enough’.  99% correct syntax in code means your code runs 0%. 1 small error in a sequence . Code is highly syntax-sensitive (and domain specific syntax at that (e.g. an API its endpoint or an SDK’s methods and events...) ; whereas a summary paragraph is not as syntax sensitive - If it is semantically coherent its acceptable… greater margin of tolerable error.
• I’m excited to see what people will do with all the extra meeting time now that they can use these tools . Miro should make an estimate and actually track implementation. Like how “loom video” tells you how much time you’re saving when you send a video.  

Miro AI Beta Testers: Let’s share our thoughts, ideas, feedback.

Here’s my TLDR; on each of the tools 

A walkthrough of my tests and my opinions on the results:

___________________________________________

Other Miro AI Beta Testers-- what are your reactions? 

I’ll publish then edit -- coming back here to type out some of the big thoughts. 

A very thorough and helpful review of Miro AI. Thank you Max! 

First off, loving this addition to Miro. I can think of all kinds of interesting use cases for the idea generator, and summarizing sticky notes. This piece of feedback is on the image generator. As a test, I asked for a map of North America, and this was the result:

Missing a few key countries like Canada and Mexico. So, I refined the search to a Map of North America with Canada and Mexico, and this was the result. I give you, Nortih Anmerdiica

The west coast has a few new land masses, and I have no idea what the island cluster is, though I would like to visit Zaplada, Cancia, Canjada, Canmada, and Canaaa. 

Interesting results. A google image search did a much better job here, but definitely not as funny. Will keep trying things out and report back findings :)

Miro AI has a monthly cap on the number of requests that can be made, what is that cap?

So happy you were able to find the form, Linda!

great can you, participate in the beta as an educator ? Thank you

Hello👋

I enjoyed watching the demo live and immediately shared my excitement with colleagues at work, but still waiting for access. I am sure at the end of the event, someone said because we joined live, we would be invited. 

Anyone else who attended, still waiting?