Discussions regarding "free speech" on social media tend to be very simplistic and often ignore psychology, sociology, linguistics, psychological manipulation, lies, propaganda, trust towards authorities, science of communication and interpersonal relationships (and power dynamics in interpersonal relationships). Many people on social media are either "pro-free speech" or "pro-censorship" as if those positions were some sort of sports teams that one must always cheer on, often ignoring the nuances and deeper implications (not to mention, there is a lot of hypocrisy and bad arguments regarding that topic, on both "sides"). This page will explore the concept of "free speech" more in-depth.
Two main positions:
1: pro-free speech
2: pro-censorship
More nuanced positions:
1: Isomorphic free speech: a situation of "free speech" (except the usual suspects e.g. slander, death threats, doxxing etc.) that is the same everywhere and in every situation.
2: Sandboxed free speech: kinda like when you try to use a bike indoors in Pokemon games. "There's a time and place for everything". That means, depending on the place and situation, what can be said varies. However, there's no top-down censorship superstructure.
2.1: Geo-sandboxed free speech: what can and cannot be said varies depending on the location (e.g. a Christian church may not allow pro-Satan speeches).
2.2: Socio-sandboxed free speech: what can and cannot be said varies depending on the situation and social context.
2.3: Holistic(?) sandboxed free speech: there's a time and place for everything, but for each speech, there exists at least one place such as that speech is allowed.
2.4: Non-holistic(?) sandboxed free speech: there may be at least one speech for which there exists no place such as that speech is allowed.
3: Isomorphic non-free speech (moderate censorship): a situation of moderate censorship that is the same everywhere and in every situation.
4: Sandboxed non-free speech (moderate censorship): mostly the same as 2.x but with moderate censorship.
5: Isomorphic/sandboxed non-free speech (high/extreme censorship): if you read the explanations above, this will be pretty self-explanatory.
Many discussions regarding "free speech" often focus on the rights of the speaker. But what rights does the listener have? (coming soon):
1: No rights at all. Listeners cannot refuse to listen (or read) a certain content if the speaker wants them to listen. How that would be enforced, especially online, is unclear.
2: Right to refuse (e.g. one can scroll away, or get away IRL), but still, it's possible (perhaps even frequent) that the person will still encounter content they don't want to "consume" (though they are allowed to scroll away or get away IRL).
3: Right to refuse + optional algorithmic filter (generated by AI?) such as a person can, through a prompt or some other manner, describe what kind of content they don't want to see, and the algorithm will hide that content from them. It's not at all unlikely, IMHO, that because of AI, in the future one will be able to describe exactly what they want in their "feed" through a prompt, and then get the content they want e.g. "I want news regarding anime, videogames, and flip phones... I don't want to see political content or videos about animals, except koalas occasionally, which should be 1% of my feed" and then the AI will generate a feed that is exactly as you described. It seems very plausible to me, and I think one will be able to have multiple feeds as well.
4: Right to refuse + right to make that content inaccessible to everyone (censor it). For pragmatic reasons, not everyone can have such right, of course, otherwise all content would be banned, but it's still a logical possibility (it's not contradictory in a logical sense to have a situation where all content is banned, just highly unpragmatic). In general, only the rulers of a society have such right.
5: Variable (due to hierarchies?). Perhaps in some situations, one is not allowed to refuse listening, and in some other situations one is.
Real life and social media are structured in different manners.
Different websites are structured differently. Some types of content can be considered "monopersonal" (created by just one person... or perhaps even multiple people agreeing to work on the same content in case it's more complex), like a YouTube video, or a story or a drawing on DeviantArt, and "multipersonal" (created by multiple people, without a strict agreement), like a thread on a forum or on Reddit, or a conversation on X/Twitter. "Free speech" (i.e. lack of strict rules) may be more problematic on a website that prioritizes content that I call "multipersonal". Moreover, different people may place different degrees of importance on comment sections of monopersonal content.