… and that particular argument is wrong. ATMOS surround sound speakers solution, which is not Tempest 3D audio’s competitor by Dolby for headphones, tries to add more height information to the soundstage with the additional speakers, but it cannot make assumptions about the listener and where they are located exactly nor about how their ears are structured (which is why companies would go and create audio profiles and then try to make configuration tools to allow you to find your closest match).
Headphones can make a great assumptions on how the sound will hit your ears, which should be obvious, and while the quality of the speakers might mean less pristine and loud sounds the sense of presence has a chance to be a lot higher. You can find binaural audio demos with shitty headphones that feel super immersive and there is many years of applied research on the topic and many companies are now jumping on that bandwagon (see Apple and their Spatial Audio tech).
Why did Sony go ahead with their solution and what is good about it? Possibly even just a Dolby patents unencumbered codec they could give away for free to PS5 devs would be a reason to go for it, easier adoption by devs and users: it requires no extra licensing for devs, not extra licensing costs for Sony, and only cheap headphones for users… on top of that they gave devs a very powerful audio chip that is fully programmable and can be used as an additional general purpose compute resource too if needed (and adapted to improvements in the audio codecs supported too).