What you must mean is a firm, not a corporation. And sure, it can be said that DPOS produces a decentralized digital firm like structure. The point is, that structure works well and allows for governance without loss of efficiency. While you can say that perhaps Proof of Work is much more automated, perhaps someday will be fully automated, there is no reason to believe Proof of Stake cannot become a fully automated "firm" or conglomerate. So the only difference would seem to be with Proof of Stake the human being ultimately has the control while with Proof of Work no one really owns it, it's sort of it's own thing.
This is the exact reason I prefer the hybrid Proof of Work + Proof of Stake approach. I think pure Proof of Work encourages long term risks which pure Proof of Stake or hybrid Proof of Work + Proof of Stake do not encourage. A long term investor who wants to know for sure their investment will be around and remain pro-social might prefer to hold the Proof of Stake project because they can always course correct in the future. In Proof of Stake some humans will always be in control as long as some humans are the stakeholders.
I mean corporation because in corporation capital ownership and direct participation in management are distinguished. Cf.: societe anonyme. 'Firm' = name / signature ... etymologically and financially and denote the 'ID' of a participant in an economy. Firm is the widest sense, it matches 'account' in the cryptosphere. On PoS as reinvention of corporation - exactly right, you nailed it - it is automation of corporation, call it so then ... it is NOT guaranteed that 'human being ultimately has the control' until there is some orthogonal to the PoS protocol ID verification protocol. Which inevitably involves off-chain ID validators. Involves TRUST, i.e. involves force. The 'hybrid' of PoW and PoS would then have three dimensions or coordinate axes - PoW, PoS and ID ver. The trick is to get to MORE efficient scheme of work, isn't it? Occams rasor cuts us otherwise. Long term investors willing higher SRI/ESG levels and public profile would rather choose outright centralized ledger which exist and are mainstream in enough electronic form, rather than 'hybrids'. Naturally 'hybrids' have worse performance vs the specialized solutions. Example.: a duck, it flies, swims and walks but flies worse than an eagle, swims worse than a dolphin and walks worse then a cheetah. Tradeoffs ... of hybridization. ( Note that I do not say that PoS as corporation-on-crypto is not possible to be better than centralized ledger corporation. But it is the end, not means to an end. The means is crypto, the end is corporation).