Yeah, that sounds familiar too, that he was using corporate machines.
Also, and it's overlooked a lot, but even people in management have no idea what is going on 95% of the time. Assuming I actually DID go to my bosses, and was like "Yeah, I want to fold@home on these machines... I'll shut some machines down at night, and fold on 1/2 of them to make up the power difference, so you'll see no difference... what is folding@home? Well, you'll be curing diseases... yeah"; it doesn't mean my boss would UNDERSTAND what I was doing... explaining folding@home to someone who doesn't know/care about computers is fruitless, there's no tax write off, there's no direct benefit from a business standpoint. I could convince them to do it since it'd be "helping the world" and stuff like that, but let's be honest: I just want those damn CPUs to fold under my name so my e-peen will grow.
If there was a distributed computing program that "counted to a really high number, then gave you points based on how high/how long it took to get to that number", promised to evolve as CPUs evolved, use SSE2, use SSE3, fully SMP, but had ZERO practical purpose... I'd probably do it. I honestly would. Because the numbers game is much, much more amusing to me. SETI@Home was fun for me since you could use different compilers: SSE2 compliers, SSE3, Pentium-specific... it was a clever, clever game of matching compliers to processors, and it was fun