Hi all 
AI Teammates react to comments in the tasks they follow. Most of the time that works well. Not always.
For example, you say something for yourself, but the AIT picks it up as it was a request.
Two problems:
- Noise. The team gets an answer nobody asked for, and trust could drop
- Cost. That run counted as a request.
What I’d like, when the Teammate isn’t confident it was being addressed:
Ask a short question (“did you want me to look at this?”) instead of running.
And don’t count that as a consumed request.
Today the default is act. I’d rather the default be check.
A human teammate would do exactly that. 
Please cast your vote on top left
2 Likes
I like this idea a lot. May I suggest that rather than having a single default, it would be useful to have a similar permission gate to how ChatGPT treats integrations? You can choose default behaviour for read/write, high risk/low risk actions, etc. This approach gives you more of a range of outcomes rather than a binary ‘sure/unsure’ → ‘proceed/ask’ check. I’m thinking this could almost be like a skill where you define the behaviour you’d like when there is uncertainty, but more structured than a freeform .md file.
Not sure this is my best explained idea and possibly overcomplicating it - but hope it makes some sense!
Otherwise, completely agree with @Arthur_BEGOU.
1 Like
Makes sense @AnAsanaUser!
Different defaults for read and write actions would be even better. I’m just starting simple
