Event details
21 Comments
Does Microsoft plan to introduce a fair-use subscription model for Cowork, rather than a purely credit-based usage system? I’m thinking of an approach similar to Anthropic’s fair-use policy, where usage is included within reasonable limits instead of being charged strictly through credits.
In which cases would you recommend someone to create their own copilot agents (included in premium license) & when NOT to?
100% agree ! Copilot pages feel like having superpowers ! Love that feature!
I have tried to use the Create function to create PPT but don't manage to get consistent outputs based on templates and there is no possibility to add a brand kit here. Are there any best practices on how to get outputs that are really aligned to my template?
This resonates a lot with me, so far, I must say Copilot in PowerPoint has improved a lot on that matter, recently, though.
agree, it has improved a lot, maybe thats why its soo frustrating when those small things don't align... The best way I have found so far is opening a PPT example and using copilot directly in PPT to add the content.
- ludoviculrich
Microsoft
check out the new brandkit capabilities in Create.
Create and manage official Brand kits in the Microsoft 365 Copilot app | Microsoft Supporthave already tried this but can't add the brand kit when trying to create the PPT unfortunately. So let's see what new thing will come here as announced :)
Hi everyone! Hi Carl, have you noticed any overlap in the step-by-step process when using Researcher compared to the Sales Agent? If so, where do you see it?
Thanks for sharing! great, I was thinking about that pre-trained reasoning language framework as well, but the idea of agents revisiting information with each other is fantastic.! super clear, thanks
Model Council sounds interesting! are there any other new and cool things happening?
This model is quite useful as well, but from what I understand, it's limited to roughly 25 interactions per month.
Another practical approach is to create an "auditor" agent that can independently review and challenge responses. If setting up an agent is too complex, an alternative is to open a separate chat, paste the original answer there, and ask another model such as Opus or GPT to review it.
That's actually how I discovered that, within the same conversation, a form of determinism can emerge: the AI tends to be influenced by its previous responses and may unconsciously try to justify or defend them instead of reassessing the issue from a completely fresh perspective. Using a second model or a separate chat often helps provide a more independent review and can reveal errors, omissions, or assumptions that might otherwise go unnoticed.
how would you best describe to users when to use copilot chat vs when to use an agent
I do not see a link for AMA questions. Will that be added later in the presentation?
CoraChen1This is the right place to ask your questions! Leave your questions in this comment section and we'll answer them as we go.
My question is: how can we access the thumbs up/down feedback submitted by users in Copilot? This feedback doesn’t seem to be available in the Dataverse ConversationTranscript table
- CoraChen1
End user - cannot access the feedback that they submitted. Tenant admiuns can acess via the Microsoft 365 Admin center - https://learn.microsoft.com/en-us/microsoft-365/admin/manage/manage-feedback-ms-org?view=o365-worldwide
Thanks for answering! My concern is this feedback is not programmatically accessible, and for all custom agents we need to develop custom feedback recording, and end users always get confused with MS's thumbs up/down.
its for custom agent.
