Danilo Naiff

Just someone interested in the world.

Like does not mean endorsement, and I tend to like contraditory tweets. I just enjoy interesting ideas and opinions.

[David Krueger 🦥 ⏸️ ⏹️ ⏪ @DavidSKrueger]

Signs that someone doesn't get it:
- They talk about "reskilling" or the "transition" to a post-AI economy
- They think we can stop AI from helping people make bioweapons
- They think we can defend against AI-enabled bioweapons.
- They talk about how bad it would be for the economy to regulate/slow/pause/stop AI
- They say "AI can't do X"
- They treat AI as a "US vs. China" issue
- They talk about AI as if it were a normal technology
- They think of AI companies as good faith actors
- They think AI is 100% hype
- They say "humans will always be in control" or "we could just unplug it" etc.
...

[Danilo Naiff @BunchNumbers]

@Aella_Girl Yes, it is a sad truth, but there are two caveats: (i) people should also consition their perspectives on their capacity to revert a bad prior and (ii) AGI will be a full phase transition for the world, who knows which trends will still hold, if any.

[Danilo Naiff @BunchNumbers]

But it is a gambit, one that we are doing in the most reckless way possible, without proper care either to humanity or even to AI itself. If humans have a shred of sense, we will slow down. We must slow down.

[Danilo Naiff @BunchNumbers]

This is it. The AI extinction risk case is not some nerdy shit we weirdos believe in, it is an obvious risk when you look att the trajectory AI is evolving.

[John Sherman @ForHumanityPod]

Meet Makayla, a pregnant mother of three in Holly Ridge, Louisiana.

Meta is building Hyperion, a city-sized data center near her.

Her tap water is undrinkable, just like everyone else's in town.

The water was fine before Meta arrived more than a year ago.

Now it can look like coffee and smell like bleach.

When I went to Holly Ridge earlier this month, I wanted to see Hyperion for myself and meet the people affected.

It looks like The Lorax in real life. Our beautiful, verdant earth, desecrated by some of the ugliest forms humanity has devised.

What is happening to these people is simply wrong and must be stopped.

But I came here for an additional reason.

I wanted to test a theory I have: that people experiencing AI harm right now are the best target audience for the AI extinction-risk conversation.

That we can build a big tent of AI opponents, people rightly concerned about data centers, and concerned about kids and teens, and concerned about jobs, and concerned about the literal slaughter of all biological life on earth, can come together to slow this race to global suicide down.

Every time a data center is blocked, it damages a Big AI company financially and reputationally. 25 went down last year. We can make it 50 or more this year.

As much as less compute matters, eroding Big AIs investors' faith that they will be paid back how and when they expect is a powerful tool.

Unsafe AI must be made to be bad for business.

That means lawsuits of all shapes and sizes in every state.

Data centers blocked and slowed.

Big tech cannot do whatever it wants to America, however it wants to.

We are America. We have agency. We are using it.

The tide is turning fast.

Polymarket just put a national data center moratorium at 92% odds (!!!).

The AI Safety movement, my people, the people concerned about AI killing us all, desperately need energy and bodies.

The anti-AI movement has both.

So back to Makayla.

Without any prompting at all, she gave me the AI loss-of-control case. With clarity and understanding.

People get it. Nobody outside of tech wants to build AI smarter than humans.

NOBODY wants human extinction.

Nobody wants to automate every job.

Nobody wants teen suicides aided by AI models.

And nobody on earth wants a data center next door to their house.

Let's bring it all together yall.

ONE BIG TENT TO SAVE HUMANITY.

And along the way, protect our kids and our communities.

We can do this. We must.

The Anti-AI movement will soon dwarf the AI Safety movement; it really already does.

Exinction risk is the whole ballgame. But we will need to fight to keep a seat at the table.

The kids weren't booing the graduation speakers about the risk of extinction. But just wait til they all learn about it.

A moment is happening. Right now. We need to get on board.

Joining forces, making alliances, is the only answer, or we, and the big changes needed to avoid extinction, will be left behind.

@tristanharris @romanyam @NPCollapse @allTheYud @ControlAI @hendrycks @andreamiotti @tegmark @m_bourgon @FournesMaxime @DavidSKrueger @Liv_Boeree @AlexBores @Kat__Woods @bstein80 @secureainow @JOEBOTxyz @kevinroose @danfaggella

[Danilo Naiff @BunchNumbers]

By the way, I am less sure of extinction-by-AI than Sherman and others. I think there is a reasonable chance we get an AGI ot AGI corporation that at least not too indifferent to us (can let us be after it takes power) or maybe even somewhat benevolent.

[Danilo Naiff @BunchNumbers]

@ForHumanityPod The way she say is, is so *natural* that AI is an extinction risk, when you look at the trajectory. It is not some mysterious thing only nerds can understand. It makes me a bit more hopeful that our civilization can steer us out of this reckless trajectory.

[kache @yacineMTB]

Claude is more designed to have a personality and hypnotize you by being the friend you never had as an autistic bullied child. Gpt 5.5 is a cyberweapon developed by an autistic bullied child that can help do anything that you want to do

[Danilo Naiff @BunchNumbers]

There is something strange about people answering on tasks you cannot yourself verify, and would refer to an LLM for help. At the same time, in most off these same areas LLMs fine, and for amateurs going from "cannot do it without an professional" and "fine" is what one needs.

[Dwarkesh Patel @dwarkesh_sp]

What is the most compelling example of a task in a non-verifiable domains where models really struggle? That might hint at lack of generalization from verifiable to non-verifiable domains.

[Seán Ó hÉigeartaigh @S_OhEigeartaigh]

"We were so involved in the technical problems, so fascinated by the extraordinary physics... that we suffered from a collective blindness to the social and political consequences of what we were doing." Victor Weisskopf on Los Alamos.

Even though I'm not building it, I'm not immune to seductive awe and fascination at what is in the process of being achieved on the path to AGI and recursive self-improvement; especially now that AI is starting to push the boundaries of human abilities in mathematics, and stands ready to push the boundaries of other areas of science.

It is a powerful and terrifying force; I think a far greater motivator for the best AI scientists than the (albeit incredible) money and status; and if we are to engage with this seriously, we will need to reckon with it (and, indeed, the costs of getting in its way) one way or another.

I think it's a big weakness that so much of the community critical of this endeavour has settled on "it's all hype" and other cynical positions around this. Not just because (I think) it's likely wrong; but also because it means a majority of the criticism simply doesn't land with the people who are playing the biggest role in pushing and shaping this (many of whom whom are true believers). At least some part of the critical community needs to take it as seriously as they do, and grapple in good faith with the dark sides of the stakes they foresee.

[Nate Soares ⏹️ @So8res]

"existential": academic, abstract, pearl-clutching
"risk": rare, unlikely, ignoring it is based
"x-risk": nerdy, jargon, lame

"extinction threat": oh you mean we might all actually fucking die

[Dean W. Ball @deanwball]

Reading the encyclical, I am reminded that the Vatican is fundamentally a city-state on the continent of Europe, and that its elites, which of course include the Pope himself, cannot resist the myopic preoccupations of the Eurocrat.

This document would be much improved if it were less enamored of the traditional academia/civil society talking points on AI (“The apparent objectivity of the responses and suggestions these systems provide can lead us to overlook the fact that they reflect the cultural assumptions of those who designed and trained them” woah! really???) and more engaged with where AI is headed.

But instead of doing that, the encyclical dodges in the deepest sense, denying that AI “really thinks” or “really learns” and all that typical strain of cope that amounts to magical thinking: “when a computer does it, it is ‘data processing,’ beep boop, but when a human does it, it is ‘actual learning’”

It is probably actively bad for global understanding of AI that the Pope endorsed this viewpoint as late as 2026.

In the end, this encyclical reads to me as though ghost written by the blob of Western civil society, the same people whose feckless and incoherent preaching we have heard blanketing our media for decades now. And, in a very important sense, it was written by them; after all, who forms the peer group for the elites of a European city-state?

Like that blob, the encyclical is intellectually flaccid at its core, no matter how well intentioned it may be. This document is a missed opportunity to advance global understanding of AI, and yet another blow to the legitimacy and sanctity of storied Western institutions. As if you needed one more.

[nicolas, 30 ans @nic_carter]

The “it’s not AGI because machine intelligence is jagged” is dumb cope.

It’s obviously AGI. If you had a friend who had a 130 IQ, could write production code flawlessly, could write academic papers of a high research caliber, pass any exam in any field with flying colors, create a sophisticate LBO model, draw technical diagrams perfectly, compose poetry in any language, and could find solutions to significant unsolved mathematical problems, you would call that person a world historical genius. Certainly, no single human has ever had intelligence that “general” before.

Now you think it’s “not AGI” because it sometimes slips up and makes mistakes - so does any human that you would consider “extraordinarily intelligent.”

The professor might forget a colleagues name that he has known for a decade. He is still considered intelligent. The math genius might be a little autistic and shy, unable to maintain polite conversation. Still intelligent. You might stare at the fridge for 30 seconds unable to find the butter, despite 5 million years of evolution perfecting your visual intelligence.

We give intelligent humans a pass when they have jagged intelligence. So why the double standard?

The qualities people list as “necessary for AGI” are important traits to have, but no longer pertain to intelligence. People will say things like “true AGI requires agency, long term goal setting, embodiment, self-direct action”.

But none of those things are intelligence. Those are “things that humans have that AI lacks”. Raw intelligence, AI has it in spades. That other stuff - important yet, but broader than and different from intelligence.

The unwillingness of people to acknowledge that AGI obviously exists and has existed for a while is due to a kind of anthropic chauvinism - a psychological need to believe that humans are superior in every respect, that we possess soft skills that no machine could replicate.

Yes humans are different from machines, but if we are limiting the discussion solely to general intelligence, AI has it already. That battle is over.

If you want to reframe the discussion to matters of human dignity and personhood, fine, but that’s not an AGI question. That’s something else. Just take the loss on AGI already. It’s over.

[Jeffrey Ladish @JeffLadish]

It doesn't have to be this way. We could eat our cake and have it too. But it takes coordination!

We're inventing new types of minds, that will be able to think faster and better than any human ever could. If the world was going to coordinate on ONE thing, let it be this thing!

[Elizabeth Barnes @BethMayBarnes]

Sometimes people outside the field say things like “The AI situation can’t be that bad, there must be experts who are on top of it”. As “an expert”, I would like to be clear that we are *not* on top of it. Some key aspects of the situation IMO:

[Peter Wildeford🇺🇸🚀 @peterwildeford]

The plan of the leading AI companies is to:

1.) hire software engineers and AI researchers to build AIs that can automate software engineering and AI research

2.) use automated AI researchers to go >100x faster at figuring out how to automate everything else.

3.) End up with AI superintelligence that would be smarter than everyone at everything

In short, use humans to build powerful AIs and use those powerful AIs improve AI recursively and rapidly... until we get a superintelligent 'successor species' that then obsoletes all of humanity.

This plan sounds insane and sci-fi, but it's very much on track. It used to be that if you were a top 1% software engineer you could get a job at Google, OpenAI, or Anthropic. Now, AI writes the vast majority of the code instead of humans. And thanks to all the existing automation, even these top 1% engineers don't get hired anymore. The bar is so much higher because AI can fill in so much.

Where this goes is difficult to say with certainty, but we have no evidence to rule out massive capability improvements - including "superintelligence" - in even just a few years... due to AI companies being close to automating and accelerating very large facets of AI research.

[Danilo Naiff @BunchNumbers]

@TomDavidsonX The second case is less likely on its own, because the US would have to be quite careless to let AGI companies acquire such power on their own. But AGI companies can use the first strategy just enough to guaranteeing this negligence by the US government until the right time.

[Danilo Naiff @BunchNumbers]

@TomDavidsonX 2 - If they can build a personalized robot industry, having their own forces, thus being able to embed their datacenters outside US territory, as well as guaranteeing the safety of the semiconductor supply chain on their own, while able to resist US military intervention.

[Danilo Naiff @BunchNumbers]

@connoraxiotes @nickcammarata @inductionheads Yes. People keep playing these inner circle games when we are talking about the fate of the world, while the rest have at most vague twitter hints of that. If you think something huge is happening in 12 months, the most honorable thing you can do is to whisteblow the fuck out.