Re: Economics are a PITA, aren't they?
I just don't even know where to start with this reply. Almost everything you said was wrong.
I'm going to ignore the breathless idiocy about post-scarcity society or AI somehow making making your meals. AI is orthogonal to that - our most advanced robots can currently do some gymnastics and make a few specific cocktails. That's a long way from cooking meals, and if you don't realise that then I pity your local fire brigade, who you will no doubt be on first name terms with soon after you get your first domestic robot.
The Wright Brothers' Flyer is a terrible analogy, because they didn't start charging people for transatlantic flights immediately after they'd first flown it.
Fifteen years after they'd flown we were flying combat aircraft in a war. The pace of development was rapid, and the constraints were clear - power, albeit of a different kind to what we were discussing. The word you'll hear time and time again when looking at failed aircraft is "underpowered". The jet engine just about fixed that after the second world war, albeit in its second and third generations.
All technologies have constraints. The problem with AI's constraint being power is that building new power infrastructure usually takes decades.
So if you want this glorious new future you're convinced is there, you're going to have to accept that it won't be quick to arrive by any stretch. And I'm going to completely gloss over the questions of the economic and environmental costs of building that power infrastructure, as that's a very hot potato.
As to the fiscal cost of AI services, I think you're underestimating how many businesses see it. I'll take a quick example for my home territory - the UK. Microsoft charge £31 per user per month for an E3 license. Which, for the Office suite plus Exchange Online and (until recently) Teams, isn't bad. But if they're charging an extra £20 per month for CoPilot (their branded AI), that's basically a ~60% price increase.
For a handful of people that's fine.
For a company of 500, that's extortionate. A jump from £15,500 a month to £25,500. Or if you prefer yearly figures, £186,000 to £306,000. Given how few use cases there are for AI right now, I'd have difficulty getting that past most finance departments...
And we're supposed to be in the "selling the razors at cost so we can make money on the blades later" stage of the business plan.
The costs are not looking like they'll work out in the short term.
I'm sure AI will change many things. But it won't cook your dinner for you without major advances in robotics, and a new power station in your area. It's lovely that you have £1000 a month of disposable income though - have you considered employing a cleaner and maybe subscribing to one of those dinner services like Hello Fresh? That might get you further towards your goals, with change to spare...