My 'Commanders Intent' for building a profitable AI business swarm:
Build an AI Automated swarm which executes business actions which add value to humans, produce profit, and self-sustain through strategy.
Broken down:
1. Executing Business Actions:
Selling narcotics could be a business action – what’s to stop the AI swarm deciding to venture into the dark web and basically form a mafia? We’ll need guide rails.
2. Add value to humans:
Quantifying the objective value of a ‘product’ to humans is hard. Humans value sweet soda, and in the short term it’s a net positive. Drink it every day though, and health is compromised. So we’ll need long-view & morality.
3. Produce profit:
Money is power in this age, and capital begets capital. The AI will need to produce more than it spend’s, but if it costs $10kpm to make $5, the risk ratio is skewed. We’ll need to specify a target profit margin.
4. Scaling:
This is perhaps the most interesting part of building a 100% AI Automated business. If the business runs itself it could produce $10 revenue, and spend $5 to achieve it. This would seem futile to all but a child running a lemonade stand.
But if a swarm reliably cleaves profit (from doing good things) however meagre or large, it potentially has the wonderful property of being scale-friendly.
I may be able to produce a hierarchy of 1000’s of niche swarms servicing tiny markets but overall producing incredible profits.
5. Self-sustain through strategy:
This is where I suspect the most frustrating work in this challenge. It’s one thing to found a business and inject your knowledge into some AI actors, but it’s another to imbue them with enough scope and wisdom to pivot, evolve, or exit when the business falters.
Read my full goals for this AI business challenge here:
https://lnkd.in/gynfz7_v
Thank you for the shoutout, dear headbits AG team!