Comment by _heimdall
10 days ago
Regulating the algorithms would be extremely difficult and I don't expect regulators would be able to do it in a clear enough way that it keeps up with whatever tech companies try next.
Much simpler would be to clarify that algorithmically curated feeds of user content do not fall under the protections shielding the hosting company from liability. It seems reasonable to me that the intent of said protections was was to avoid scenarios where the website is in no way directly involved in the content creation itself, once the company decides what and how to show it that seems outside the original intent.
Yup, critics of Section 230 often cite the possibility doing so will chill speech, that companies will start taking things people say down, to avoid any chance of liability.
Perhaps instead of putting bounds (through laws) on what they cannot do, perhaps we list only what they are allowed to do (for algorithmic feeds), eg. can only show content from direct connections in chronological order. I do expect many people (currently believe) they like being able to discover new content outside of their network.
Whatever we come up with, which would benefit the people, I suspect will have big impacts on their profits. At the same time, if they do not have to spend vast resources to understand our nuances and target ads specifically to use, their cost centers would also be impacted.
I'm a growing proponent of ending targeted advertising. If we start with that, much of the incentives for our current algorithms (social, search, news) would seem to be removed from the equations.