this particular case really seems like one of the easier cases in which to reject the request, which they would probably win if litigated.
Which is exactly the problem with this law -- if Google wants to reject a request, the burden of proof is on Google to prove that the information should NOT be deleted. While you may believe Google could win this case, you're suggesting Google spend legal resources (and money) fighting over a few links... and that they should do that every time they get a bogus request. AND if they repeatedly do this and lose their case, Google may be subject to fines of 2% of worldwide annual revenue (over $1B/year).
Why would Google ever decline a request when the penalty is $1B/year?
That's certainly a possible concern, but I'm not sure it's the one in question here. Is there much chance that: 1) O'Neal would actually sue; and/or 2) win? My guess is that if Google rejected his request, he wouldn't sue. Both because of the bad publicity, and because he's in a fairly weak position to invoke this right with respect to his past as CEO of Merill Lynch, a very public role.
I think the bigger issue for Google isn't their fear of fines and/or losing lawsuits if they really did make case-by-case determinations, but rather than difficulty & expense of putting together any kind of system for case-by-case determinations in the first place. They tend not to want to put together such systems, as can also be seen in the "delete first, ask questions later" approach to letting music labels delete videos on YouTube after algorithmically scanning, not even requiring the labels to put together a proper under-penalty-of-perjury DMCA request. I'm sympathetic to that position from a practical perspective, but I think it's a somewhat different problem. The problem here is that even if it is perfectly possible to make case-by-case decisions in a reliably compliant way, Google just doesn't do case-by-case decisions in that manner, for business reasons: their entire business model is automated decisions at scale, not manual curation of content.
Which is exactly the problem with this law -- if Google wants to reject a request, the burden of proof is on Google to prove that the information should NOT be deleted. While you may believe Google could win this case, you're suggesting Google spend legal resources (and money) fighting over a few links... and that they should do that every time they get a bogus request. AND if they repeatedly do this and lose their case, Google may be subject to fines of 2% of worldwide annual revenue (over $1B/year).
Why would Google ever decline a request when the penalty is $1B/year?