

One of the major draws of discord is the fact that they host the servers for you, for free. Anyone can make an account, click a button, and have a discord server.
Afaik matrix does allow this (haven’t used it personally) but it’s something where I am a bit worried about hosting costs if it reaches a large scale. (Also unsure about how the matrix protocol works precisely, but if defederation is a thing which I feel like it has to be, I can see it leading to huge pains since discords use case is often about being part of a specific communitu, as opposed to twitter or reddit. Being unable to join a groip or see some messsges because of federation issues would be a major headache).
It’s less of a bias of the programmer and moreso a bias of data, particularly when a factor like gender or ethnicity correlates with something without direct causation, such as crime rates correlating with ethnicity largely because of immigrants being poorer on average, and economic standing being a major correlating factor. If your dataset doesn’t include that, any AI will just see “oh, people in group x are way more likely to commit crimes”. This can be prevented but it’s generally more of a risk of overlooking something than intentional data manipulation (not that that isn’t possible).