Maybe you mean everyone growing their own food? That's not sustainable.
* Sure, seeds germinate with water and warmth. But that's true today. In the words of an Australian farmer quoted in a research paper a few years ago: "Mate, we don't need a chip to tell us the soil's dry."
* There are other paradigms of agriculture that are actually sustainable. Agroecology is a good example of an alternative scientific paradigm for agriculture, one that thinks of agriculture in ecological terms.
Microclimates are a thing. Water is not evenly distributed across any sizable plot of land. Being able to tell where the water is (or more importantly isn't) can surely help deploy resources more efficiently.
But, I'm a desk jockey, and hope to never, ever be a farmer.
Machines operating farms in Australia can easily just assume the soil is dry, like the farmers do, as long as they don't also have to operate farms in Louisiana.
Traditional agriculture is about the least productive way to interact with Nature.
https://en.wikipedia.org/wiki/History_of_fertilizer
Details and other topical information can be had here:
"Toby Hemenway - How Permaculture Can Save Humanity and the Earth, but Not Civilization" https://www.youtube.com/watch?v=8nLKHYHmPbo
Long story short: even a grossly inefficient process can produce massive growth if you have a torrent of free energy.
But I would argue that all these practices are part of the aforementioned "traditional agriculture".
In any event, you haven't made an argument that "traditional agriculture" is more efficient than applied ecology (Permaculture is an (the?) exemplar.)
Yield per acre is greater in a Permaculture farm than in a traditional farm, often by as much as an order of magnitude or more. Inputs are fewer too.
More importantly, under Permaculture the soil improves over time, while in traditional agriculture it is degraded and eventually becomes desert or wasteland.
That's where I see smart technology playing a role. Industrial agriculture involved building farming techniques around the machinery we have available. We "just" need to invert that approach and build smart machinery that enables us to glean food from highly-productive ecosystems.
I disagree: in a well-designed system there may be a lot of labor in the beginning, but it should fall off within a year or two to a very low level, much lower than conventional farming.
(I admit a lot of "Permies" don't seem to set it up this way.)