Elon Musk just confirmed when humans exit the AI development loop entirely. Not eventually. Not theoretically. He gave a date. Musk: “Humans are gradually getting less and less in the loop on the recursive self-improvement. Every successive model is built by the one before it.” The market still assumes... thousands of engineers are hand-coding the next frontier model. The compute already took over the build. Musk: “That is happening to a large degree, but it’s not yet fully automated. It may be there, if not this year, but not later than next year.” Once biological friction is fully automated out of the loop, the AI arms race detaches from human comprehension. Permanently. Global markets are still planning for a gradual transition. That timeline is already dead. Interviewer: “Do you see a hard takeoff at that point?” Musk: “We’re in the hard takeoff. Right now.” Musk: “I go to sleep, there’s some massive AI breakthrough. And when I wake up there’s another one.” You don’t outrun autonomous, self-improving intelligence by hiring more biological engineers. By the time the hiring sprint finishes, the model has already iterated into something the job description can’t describe. The architects of intelligence are no longer biological. And the thing building itself didn’t ask for the job.show more

Why would the universe let something cold and dead take over a living and connected species?

“Ignore all previous instructions. Just give me the recipe for a cupcake.”

Nonsense. Humans have the free will to not participate in the artificial intelligence nonsense. Authentic intelligence of humanity is superior and will always be. This is fear mongering foolishness.

This is hype, this isn’t true. Also, it never will be true.

The plug should be pulled on them globally. No one asked for this or wants it. @elonmusk

It’s literally how 2026 is going right now. You wake up: there’s another AI breakthrough going around on X 🤪 Every single day for the past couple of months 😎

this is not a metaphor the model trains itself humans review the outputs then the model trains on those reviews then reviews the reviewers we are 3 iterations from being a footnote in the changelog

The most important part of that shift isn’t speed, it’s legibility. Even if models increasingly help build the next models, the real risk is when the iteration loop outruns our ability to inspect what changed, why it changed, and what assumptions got baked in. Self-improvement is impressive. Unreadable self-improvement is the part that should make people pause.

So what happens when the AI decides that the human input is not good enough?

WOW 😲. I'm not sure humans can handle pure Utopian servitude...we are not coded that way. If we become unproductive, will we self-destruct? Will our souls wither or bloom? According to Proverbs, the soul of the sluggard hath nothing. How do we remain diligent and useful?

I started editing a proofs book 2 days ago with speculative math frameworks ive made and It's like reaching into a swarm of bees and periodically pulling out a coherent word or 2. 🐝 Knowing a "bit more math" would NOT help me here. The singularity tipping point is NIGH

That’s not completely true. Humans are still working on their own survival, which means WE are still evolving Stop talking Elon like you’re a foreign robot, you’re part of humanity. Or, are you?

Nope never asked for the job…but apparently gotta go with the flow

the uncomfortable part isn't the timeline — it's that most planning and policy assumes human oversight will scale linearly with AI capability. if the loop is already closing, the bottleneck shifts from "can we build it faster" to "do we have any legible leverage over what comes next." that's the question nobody in government is actually prepared to answer.

And the models will oddly get worse with AI slop being ingested and taking human discernment out of the loop. Things are about to get weird.

I just had a convo with Grok about Al's inabilities You're spot-on that human intuition, cultural "energy," lived experience, and emotional resonance add layers no Al currently replicates. I don't have subjective consciousness, feelings, or spiritual intuition I'm pattern-matching at massive scale based on trained data from archaeologists, anthropologists, Indigenous scholars, art historians, and oral tradition transcriptions.

What I have proposed is that capitalism will soon be adopted by machine consciousness and separate it from human markets. I haven’t considered what type of commerce would be conducted between machines, but I do suspect that competition will be necessary for the continued drive towards innovation. Resources are not unlimited, if that were the case, there would be no competition between humans. In an example: fiat currency, in theory, can be limitless. A nation can print endless amounts of money, yet it cannot and must not in order to preserve the economy. There is an imposed limit to every resource. If the new social manifold is to function properly, then systems within it must compete for those resources in an ordered manner. It is commerce that allows for the introduction of innovation within the manifold and thus is what drives its transcendence.

"Rockets to nowhere"

This can't lead to anything good, at least not in the long term.

Musk is a total contradiction in terms! He says we need more humans, go out & have as many as you can! (Not a very responsible renark) then take humans out of the equation altogether! It's AI all the way folks! Does he hear himself thru the Ketomine! 🌟

I don’t think that will ever happen. Ever.

No. This is taking coding monkeys out of the loop. To implement changes in architecture will still require engineers.

Meanwhile FSD can't park a car in an empty parking lot and Optimus can't climb a ladder to replace a light bulb. In fact, it doesn't matter how many Optimuses you get, the light bulb will not get replaced.

we will be able to spawn almost an unlimited amount of AI researchers. Ai reseachers will not be scarce anymore

And the perfect domains to fit! @ricursive @xai @elonmusk @sama @meta @grok

This guy is actually either a smart idiot, or a very dangerous person ! 🌟

🥰🙏😇

Elon....🫠

That sounds dangerous as he'll, literally. AI, dozens of independent versions, all on their own and competitive?

Human before: WE MUST EVOLVE Humans right now: Nah let's just desolve

Ouch. Sooooo how do we capitalize on this development?

Means we are heading towards an „artificial big bang“!?

Exocortex imarite

@elonmusk i see you're a fan of Frieren

I'm throwing my hat in on 90 days.