If you're assuming that we have the technology now in its prototypical early-adopter phase then your estimate of 2030 is for adoption of LLMs and generative AI.

We don't have even prototypical AGI yet, and it is AGI which is necessary to "do all the work." The s-curve of technological adoption assumes that the technology exists at the start of the curve. This is what I meant when I said "I really don't think we are anywhere close to this."