Irresponsible is an adjective that means lacking a proper sense of responsibility, or acting without thinking about or caring about the potential consequences of one's actions.
They do not care if they waste everyone’s time or flood the common with slop. They did not spend their own time to verify that the Lean proofs correspond to the natural language proofs. They did not even spend their own time to verify that all of these proofs are well written.
They instead are mining the unrenewable resource of open math problems.
However, seeing it another way is easy if you are financially motivated by their upcoming IPO (see how easy it is to invent motivations for comments?).
Actually they sat a highly illustrious group of mathematicians down to advise them before releasing -- illustrious enough that I am certain that they have more global standing than you do to make a good call. That group advised open release, which is what they are doing. It lets mathematicians interested and skilled in each area look at what we have. It is by no means irresponsible.
> If AI labs produce significant mathematical results, they should responsibly release the results, as outlined in this document, as soon as possible.
And
> only the mathematical community can undertake the assessment that is needed.
Making this work public is a first step. This release is the beginning, not the completion, of the process of human understanding and the incorporation of the work into mathematical knowledge.
You invoked the idea of political/legal/regulatory action to stop this type of thing. That's not commensurate with the type of irresponsibility you're describing i.e. publishing slop and wasting people's time.
What you suggested is however commensurate with some sort of personal animus.
Publishing a math paper and then unpublishing it is not "irresponsible". It's just a math paper.