AI-generated summaries of long comment chains

Is your feature request related to a problem? Please describe.
This feature is related to the problem that long comment chains create, both for participant readability and performance (as captured by https://github.com/decidim/decidim/issues/12578). A highly active decidim instance can generate over a hundred comments per proposal, on multiple proposals, which creates equity issues for people who don't have the time to read all of them.

Describe the solution you'd like
For comment chains above a defined threshold number of comments (e.g. 8), a new text section "Comments summary" would appear in between the body text (e.g. proposal, meeting, etc.) and comment chain that would be an AI-generated abstract.

The summary could include direct links to comments that appear more authoritative (i.e. linking to external sources, thumbs up feedback from participants, non-negative sentiment analysis), which would be an incentive for participants to write high quality comments.

The abstract would be updated any time a new comment is added, when above the threshold number of comments.

Does this issue could impact on users private data?
Ideally the AI used would be open-source based, to keep data entirely within the organization, but most or all of the data being processed would be public and therefore not impact private data.