I think what trips people up sometimes is they are conflating "easier" with "more powerful"; when I first read the Ali guy's stuff that's what I thought they were trying to say or at least, that's how I interpreted it, but given what we know about both systems and can extrapolate will be further features and capabilities in both of them, I honestly see them as two different things.
I.e a system doesn't need to be "more powerful" in order to be easier to develop for. The Dreamcast was relatively easier to develop on than the PS2, but outside of some select abilities (lack of deferred rendering, VRAM amount, native output resolution support etc.), it was certainly more powerful than Dreamcast. It's kind of similar with PS5 and XSX; there are some areas PS5 may have some advantages in, but for the main grunt of things and overall the XSX is the more powerful system.
However, some of PS5's advantages, namely "unified" memory pool (actually both systems have unified pools, there are no "split" pools going on in XSX but the bandwidths differ between the 1 GB + upper bound 1 GB partition of 2 GB chips (560 GB/s) and the lower-bound 1 GB partition of 2 GB chips (336 GB/s)), faster GPU clock and more powerful flash memory controller at the hardware level, are a good bit more "automated" for developers, generally to the point where they don't need to think about them too much. Not to say they don't have to think about them at all; if a game's particularly taxing on the GPU power budget due in part to the fillrate (which is influenced by the clock speed of the GPU, i.e it's not something a dev needs to explicitly tell the GPU to do through code), they may have to cut back some of the visual complexity of the scene (as one example). But compared to leveraging some of XSX's hardware headroom and hardware/software advantage potential, it could be easier to do that type of stuff on PS5 (requiring less dev "thinking").
Personally I think the XSX has more potential for "programming creativity" over PS5, particularly after the first couple of years of the systems being out. I don't mean that in terms of game design or such because both systems are capable enough to realize practically the same conceptual ideas in practice as the gen gets underway and as it progresses. However, I think there'll be more "room" regarding XSX for a bigger base of optional approaches to reach similar results, some of which can possibly prove more efficient on resources (both hardware and software) than would've been initially assumed, versus PS5 where going with certain fixed hardware functions somewhat limits what type of options can be explored in this reaard.
This is a really rough example, but think of the SEGA 32X and SEGA Saturn. Both systems could do 3D and Saturn was a lot more powerful at it comparatively, but since the Saturn used fixed polygonal hardware it could only draw polygons as quads. Conversely, the 32X could do both quads and triangles mainly due to that lack of fixed polygonal hardware, though it needed to use the CPUs to do such through software instead. Given how devs came to prefer triangles over quads, that ironically gave the 32X an advantage over the Saturn even if it came at a hardware cost.
I'm not suggesting the "programming creativity" between PS5 and XSX will be at that scale (you couldn't do a straight comparison anyway which is why I said it's a rough analogy); just suggesting sometimes easing up on dedicated fixed-function hardware and having flexible hardware headroom in one or several areas can invite certain approaches to programming and design challenges that can meet results comparable (and maybe at times even superior) to what the fixed-function hardware can provide.
Anyways, we've got a whole upcoming gen to see how all of this shakes out and that's got me kind of excited.
A scale-me-down. Something I'm not personally interested in, either, but it won't be holding back XSX from a technological or game design POV.
From a manufacturing, pricing and marketing POV, though,...well that's honestly still up in the air and the main reason I'd rather it wasn't a thing, at least not for a couple of years. We'll find out soon, though.