Dr. Roman Yampolskii literally created the term "AI Safety" 15 years ago. He's a computer scientist who's been warning about this before it was cool. For the first 5 years, he was convinced we could solve it. Then he realized: "Every single component is impossible to solve."

His prediction timeline will shock you: 2027: Artificial General Intelligence arrives 2030: Humanoid robots replace physical labor 2045: Superintelligence beyond human understanding "We're looking at 99% unemployment. Not 10% - ninety-nine percent.”

But here's the terrifying part: "While progress in AI capabilities is exponential, progress in AI safety is linear. The gap is increasing between how capable systems are and how well we can control them." We're building alien intelligence faster than we can understand it.

Companies add restrictions like "don't swear, don't say bad words" - but these are just patches. "If you're smart enough, you always find a workaround. You push behavior into different, not-yet-restricted subdomains." It's like HR manuals for superintelligent beings.

Here's what broke my brain: "We cannot predict what a smarter-than-us system will do. By definition, if it was something you could predict, you would be operating at the same level of intelligence." It's like your dog trying to predict why you do a podcast.

Sam Altman gets DESTROYED in this interview. "He's gambling 8 billion lives on getting richer and more powerful. Some people want to go to Mars, others want to control the light cone of the universe." "We published guardrails for AI development. They violated every single one."

And the unemployment reality is brutal. Yampolskii asked his Uber driver: "Are you worried about self-driving cars?" Driver: "No one can do what I do. I know NYC streets like no AI." "But we already have self-driving cars. It's not IF, it's how soon you get fired."

Even creative jobs aren't safe. On podcasters like Steven: "AI can read everything you wrote, train on every podcast, optimize better than you because it has the data set you don't." "Visual simulation is trivial. Generate videos of you interviewing anyone on any topic.”

But the most chilling revelation? "Training AI models requires vast amounts of data. Companies are valued at billions claiming they can solve this. We haven't seen their output yet." "Creating perfect safety for superintelligence is impossible. Not difficult - impossible."

The simulation theory connection: "I'm pretty sure we are in a simulation. The moment advanced AI becomes affordable, someone will run billions of simulations of this exact moment." If we can create human-level AI + photorealistic VR, someone will simulate everything.

What can we actually do? "Our job is to convince everyone with power in this space: You're doing something very bad for YOU personally." "If people realize building this is bad for them, they won't do it.” Translation? We can decide not to build general superintelligences.

While Big Tech races toward post-human superintelligence, there's massive value in serving the humans who want to remain human. Local AI, community-owned tools, creator-controlled platforms. The market for human-centered technology is about to explode.

But here's the harsh reality… Most people working on AI have no background knowledge. “They never read a single paper by anyone." "They hear 'AI will take over' and think 'it has no hands, how could it?' They don't engage with the actual arguments."

The race to superintelligence is insane: "If aliens were coming to Earth in 3 years, you'd be panicking. But most people don't realize this is happening." "We're creating technology that impacts other people without asking their permission.“

The window for changing course is closing fast. As Yampolskii says: "We as humanity are about to lose our jobs. What do we do financially? What do we do for meaning?" The answer isn't just technical - it's building communities that put humans first.

This is exactly why we built Anarchy. A platform where: - Community rewards creators, not exploits them - You discover new tools before they trend - You only pay for the credits you use & never lose any Join & try our free text to image generator today: https://anarchy.build/

Anarchy is founded by @t_mdontplay & @tom_humber. We’re building the go-to AI hub for creators & consumers to win in this new era. Follow @anarchy_build to stay ahead of the curve.