The popular belief that AI is a direct existential threat to humanity is wrong, though it’s not entirely without basis.
Let me clarify.
While many worry that an artificial superintelligence will Bostrom its way to a soulless universe turned paperclip factory, this is unlikely given the current direction of AI.
As long as AGI and ASI rely on LLM systems as their central decision-makers, we can be reasonably sure there won’t be any runaway AI deciding humans aren’t worth considering in its long-term plans. These systems won’t have minds of their own; they’ll operate strictly within boundaries set by their creators.
Which is good. But also bad.
And this is the paradox we will live in. On one hand, the fact that AI stays within confines set by its creators is a win. It prevents dystopian AI-vs-humanity scenarios and keeps us safe. We’ll enjoy the increases in productivity and new technologies, feeling relieved that there’s no rogue intelligence out to end our existence. But that same safeguard means full control remains in the hands of those who design and train, align and bias these systems. This means we’ll also see how the leading AGI and ASI systems are increasingly beneficial to the few who own and manage them. They decide how the AI is used, who reaps the rewards, and how society shares the spoils. So while our worst fears don’t come true, the price we pay is a new form of dependency, with vast power flowing upward rather than dispersing evenly to the rest of us.
As AI and AI-remote-controlled robotic avatars handle the bulk of labor, the global economy might enjoy a golden phase of unprecedented growth. Governments can use taxes on sudden profits to balance things temporarily. They can then use this money to fund social programs or a universal basic income, paid for by the owners of the AI systems. People might feel momentary relief: mundane tasks and routine monotony vanish, replaced by automation to our endless convenience.
But as this shift continues, the few who control these AI systems and robotic fleets deliberately and quietly consolidate their hold on core industries. The Cantillon effect, but for productivity and on steroids. Over time, they lock down all critical manufacturing, all global transportation and all communication networks. Even defense forces and armies transform into tech-driven units, using automated weapons and drones managed by a few centralized server farms.
Eventually, only a tiny number of individuals will hold sway over the planet’s resources. Their success won’t depend on widespread consumer spending or broad competition. Fresh startups or small innovators will find themselves boxed out — any attempt to build a new product or service demands access to the same AI networks and robots owned by the few. In the end, there’s nothing for the rest of us to leverage.
No giant labor force to revolt, because manual labor has been replaced by automated avatars. When there’s no significant human workforce left to organize or strike, the idea of a labor revolt collapses.
No broad consumer revolt either, as consumer choice won’t matter for global production that adjusts instantly to new market conditions. That’s assuming our demand even dictates market conditions, which is the best case scenario. There is an even greater risk that an economy managed by centralized AI primarily produces specialized goods or resources for a small elite. As long as their needs are met, there’s no pressure to cater to the broader population or worry about consumer boycotts.
And finally, a forceful revolt is off the table, since automated armies and AI-driven defense networks answer only to their creators, making any conventional rebellion impossible.
The masses become an afterthought in an economy that needs them less every year, with no real way to oppose the concentrated power of AI-run empires. That’s why I call this an “economic singularity” — it’s a tipping point where all global wealth and power become irrevocably centralized in a single point, reshaping our social and economic fabric.
We may see attempts to regulate or introduce new laws to safeguard public interests, but these might be overshadowed by powerful corporate and government alliances. Meanwhile, questions about ownership, freedom, and personal agency will grow sharper. In the past, humans could switch professions or find new roles whenever technology disrupted certain jobs, but in a scenario where every profitable activity can be instantly automated, there is little room for people to adapt. What role will the rest of us still have? Will we get to decide that, or will we be forced to accept whatever is left?
If we do nothing, we risk losing our remaining power. If we act now, maybe we can still shape how this new world looks. The time to decide how we coexist with these colossal AI-driven structures is here, and it’s the only real choice we have left.
