Marketing tip: Always ask the public for input and consult with "experts" before doing what you intended to all along.
Almost like act first, ask permission later.
I would like to point out that regarding moderation, tech companies have been asking for guidelines and new laws for a decade now. Ageing lawmakers have been unable to provide them and we start to see shy initiatives from the EU regarding privacy laws but roughly most of the water in which tech companies operate are still no mans lands.
All that make me think that even tough you're right that the end result will not be democratic, I wouldn't throw them the stone and accuse them of acting in bad faith.
Twitter can and do ask for user feedback all the time, without making a press release about it.
I don't find it cynical to ask why they are doing this in public - it seems likely that the parent has hit upon a good part of the reason.
If one company takes a more principled approach to moderation, and that more principled approach is harmful to revenue, that company will be outcompeted by companies with whatever moderation policies optimize for engagement / revenue / growth. As a result, in the absence of legislation, you get adverse selection i.e. the dominant platforms will be the ones that optimize for engagement/revenue/growth, rather than the ones with good moderation policies.
If you instead have legislation for what "good moderation" looks like, it applies equally to all companies and mitigates the adverse selection problem.
Of course, it is still entirely possible for bad legislation to introduce worse problems than the adverse selection problem. It depends on the object level of what exactly the legislation is, rather than being a blanket "legislation bad" or "legislation good" sort of thing.
They do have a rationale. The issue is that it's biased.
EDIT: The above are mostly something I try to keep in mind with government public consultations. I acknowledge that cynicism is a little more understandable for private industry, as they are not even in theory rooted in anything but shareholder value-maximization, but gov is at least ostensibly aspiring for public good.
Consider if Biden dropped a tweet saying "don't observe DST."
That could be dangerous, right? You've got two distinct groups of people that for some reason or another fail to coalesce, all "hell" breaks loose because some people are showing up to work at different times, logistical break down... Do you squelch him? And how does that compare to a non-violent break-in at the capitol to show congress that their representation of the general public is failing a given demographic?
Personally I think that the channel should be wide open to anything but the most unbearable aspects of communication, like child exploitation. You've got no chance of convincing a white supremacist that his worldview is askew without engaging in a conversation and logically defeating his assertions, not to mention the fact that publicly shaming the ideology (while jointly defeating it) in a widely available forum is certainly the best prophylactic. Instead the "acceptable" rhetoric is one-sided which alienates anybody that has the audacity to even ask the questions proposed, while destroying their ability to come about a rational conclusion through empirical observation. That channel should be equivalently wide for politicians which the people have elected to represent them up to the extent of tangible action to break the law up to the barrier of reason, e.g. meaningfully inciting violent actions against individuals or groups.
And this is all precipitated by the fact that Twitter and the like are commodifying speech, which is a genuine hazard as it creates a serious hurdle at the intersection of liberty and commercial interest. Commercial interests want inoffensive discussion which appeals to the widest possible band of individuals, meaning that the content and discussions are only allowed to span a narrow width, generally. It is not in any meaningful way acting in the public interest at that point, and only seeks to, through largely automatic processes, extend and crystallize the status quo which is genuinely harmful as we're doing little more than discussing how to spin in place at that point. This is driven even further through deplatforming (active or passive) extremist viewpoints, and exposing them disproportionately with algorithmic processes.