I've been dealing with this issue in my organization since one month ago.
We blocked read.AI in Teams Admin center, in web-filtering and in our anti-spam solution.
But this still doesn't prevent third parties that are coming to our meetings, from inviting read.AI agent into the meeting. If participants are not aware and admit read.AI to the meeting, it will be there recording everything. And this is still a problem we try to avoid. For now the quickest solution is to kick read.AI out of the meeting by writing 'opt out' in the chat.
I am in the EU area and I believe read.AI is mature enough to get supervision from the Data Commissioner's office. Their business model is very intrusive and it totally ignores privacy.
So please everyone please keep posting your issues here so we can have proofs to share it further with authorities.