Social media's business model is "selling ad space on a conduit that people are attracted to precisely because they can post whatever they like". There are obviously legitimate uses of such a facility. Equally obviously, there are illegitimate ones. If you can't filter out the latter, then your business model is unimplementable within the law.
Sorry, but society doesn't owe you a business model. Find a new one that you can keep legal.
I note, for example, that most B2B scenarios would have fairly good authentication of who is posting and so action could be taken against those who abuse the facility and in a business setting that is probably enough of a deterrent. Amongst the general public, however, the authentication is almost nil (how easy is it to create a new and basically anonymous account?) and even if you can identify an abuser, the likelihood that either you or the platform can take action against them in any legally useful sense is almost zero. IBM's defence of B2B use-cases is therefore more than just defending their own turf -- it actually makes sense.