If AGI is around the corner, I don't believe one single company will "own" that tech. It will be like it is today, where you have multiple models competing.

And after that, AGI will be open source.

In the end, ownership of data and compute will be the things that define the victors.

That would be true if the product was the goal. In my experience,

  marketing > market > product

Even with AGI in hand, there will still be competition between offerings based on externalities, inertia, or battle-testedness, or authority. Maybe super-intelligence would change the calculus of that, but you'd still probably find opportunities beyond just letting your pool of agents vibe code it.

If you believe AGI is round the corner (I don't), then you face the dilemma of investing in a company that will be the one profitable survivor in a disintegrating world.

This would only be true if it was cheap to run and would return results quickly. If AGI only has compute to serve 1 customer per hour then their is an upper bound of market share it can take from other products.

So grifting for investor cash and revenue right now is the obvious play either way

We don’t eat intelligence: bikes, bottles, food, energy… have room for improvement but I hardly see how AGI would replace them.

Same for physical services like labors, miners and cooks, even taxi/bus drivers for +99% of the world. Automation immensely improve their efficiency and the Modern Times is the past for half of the globe, but AGI isn’t the main facilitator.

Replace all (most*) Silicon Valley -and cousins- similar "products" and services, perhaps yes !