I think this is brave for him to say, and also partially correct.
As far back as 2003 Don Davis (himself a trumpet player) was layering synth brass with his recordings because what he was writing just demanded too much precision. Hans Zimmer had his assistants go through the string ostinatos for Pirates or Dark Knight (I forget which movie this story is about) and move them note by note onto the click - which you could consider either "resampling" or "quantizing live musicians."
If you go back and look at the raw recordings for ET or Indiana Jones these are the top session players in the world operating at a time where composers like John Williams put far more challenges on players's musicianship and technical ability. And the recordings are wonderful and lively but they're also really messy. You can kind of tell they are sightreading and there's a big contrast to any of the polished concert-suite recordings of JW's scores.
Rather than living in a world of "Live vs Fake" I think it makes more sense to shift to a realistic appraisal of what live orchestra can accomplish. There are three ceilings on what you can get from Live. The budget for quality musicians and a quality hall; the collision of limited recording time with musician-challenging writing; and perhaps increasingly significantly, the fact that what composers are writing is straying from what is natural and idiomatic for the instruments. If you are someone like Mike Verta then this is a tragedy that everyone's trying to one-up each other with unplayable 10 second long fortissimo brass chords. But IMO this invites the reappraisal that music that was written for virtual instruments should stay virtual.
Right now we're in a world where you simply cannot write a score without either needing to be a very capable VI performer, or hiring a synthestrator to do a pass on all your music, or possibly both. ("Whistler" composers now need a virtual orchestrator which combines the roles of orchestrator and synthestrator). The point is that once you have that expense paid, the virtual score is kind of there for free. Why spend more money to do the whole thing again live? The better VIs get the more the mindset of "where can live add relative value" makes sense. For solo winds and solo brass, virtual instruments fall flat and will continue to fall flat until the VI market welcomes the existence of VIs which are several times more expensive and complicated to learn and perform. Solo strings, as much as I hate to say it as a cellist, I think we're getting close to acceptable solo VI strings. The practice of "Sweetening" (re-recording just one or two elements on a track live) is indeed not very glamorous. It's not something a producer can make a behind-the-scenes video about as Craig Safan alluded to. Often it's remote musicians who aren't even in a real recording stage. But all of these things help add value because when you don't have to pay for a stage, you don't have to deal with recording time pressure, but the one sweetened element helps the whole track feel real, you're getting a lot of value for money.
The other area where complete reperformance, not just sweetening, is IMO a necessity and will continue to be, is any kind of live breathing music with fluid tempo.