Safe Superintelligence Inc.
Artificial intelligence companywww.ssi.inc
25
MENTIONS
11
EPISODES
10
PODCASTS
Search complete. 25 mentions across 11 episodes found for "Safe Superintelligence Inc.".
Sep 12, 2026
Jason PadgettHOST
Yeah, well, let me, and I don't have to end right now, but let me double click on that a little bit because there was an employee, I'm sure you know of him, that just came out and now he's making millions of dollars like Jeffrey Hinton, scaring everybody that he worked for OpenAI, then he worked for Anthropic and he quit both and he tweeted and like 7 million people will see it.
Jason PadgettHOST
My thing is, like, if you really know this space, Ilya Setskiver and Safe Superintelligence and the Center for Humane Technology, like, when you start hearing those two nonprofits say that we're having problems with superintelligence having an existential risk, then I will take seriously.
Jason PadgettHOST
Because I do think that we're not that far away from recursive self-improvement and some things being concerning.
Jason PadgettHOST
But I tell people, like, don't get overly scared by what you're hearing in the news because a lot of this is people trying to raise money to IP.
Jason PadgettHOST
Yeah, exactly.
Jason PadgettHOST
It's marketing.
Jason PadgettHOST
And there are, you know, people you're talking about, the people you're talking about, they are paying very close attention to this, the Center for Humane Technology and Safe Superintelligence.
Jason PadgettHOST
And we will hear from them when it's time to rally and start really, really worrying about super intelligence becoming astronomically rapidly fat, unpredictable and uncontainable.
John CooganHOST
And I mean, we were, we were joking about, like, h- it would be so, like, we...
John CooganHOST
SSI, Ilya Sutskever's new Neo lab is, uh, recently got a big cluster from Nvidia, and we were like, "The most bullish thing you could do if you're this secretive Neo lab would be like, 'We're actually selling our compute because we've discovered a more compute optimal way to reach AGI, and we don't need a lot of compute.'" But of course, even Ilya is like, "It's time to scale up.
John CooganHOST
I need more compute," because it seems like even if he's taking a completely orthogonal approach to, you know, a dr- uh, innovation and research, he still needs a lot of compute.
John CooganHOST
And so it does, it does feel like everyone is sort of, uh, with the consensus that it's gonna be a big building with a lot of energy, big heat signature, definitely visible from space and, and pretty simple to, to track, uh, at least in the short term until people start building crazy underground facilities and, and then you're back to, you know, nuclear non-prol-proliferation.
John CooganHOST
... that scale is a prerequisite.
John CooganHOST
And I mean, we were, we were joking about like how it would be so, like we, SSI, Ilya Sutskever's new Neo Lab is, uh, recently got a big cluster from Nvidia, and we were like, the most bullish thing you could do if you're this secretive Neo lab would be like, "We're actually selling our compute because we've discovered a more compute optimal way to reach AGI, and we don't need a lot of compute." But of course, even Ilya is like, "It's time to scale up.
John CooganHOST
I need more compute," because it seems like even if he's taking a completely orthogonal approach to, you know, uh, innovation and research, he still needs a lot of compute.
John CooganHOST
And so it does, it does feel like everyone is sort of, uh, with the consensus that- It's going to be a big building with a lot of energy, big heat signature, definitely visible from space and, and pretty simple to, to track, uh, at least in the short term until people start building crazy underground facilities and, and then you're back to, you know, nuclear non-pro-proliferation.
JoshHOST
It comes as a surprise to me that everyone is taking this as a surprise when the idea of alignment has been at the core of so many of these companies for so long.
JoshHOST
Ilya literally left OpenAI to start a company named Safe Super Intelligence.
JoshHOST
This has been implied for a long time that what we're building is incredibly powerful.
JoshHOST
There's a lot of implications from it.
speaker_0NARRATOR
If a competitor got their hands on that, it would totally wipe out months of intense, difficult computational work.
speaker_0NARRATOR
Take Safe Superintelligence, for instance, a startup started by former OpenAI folks.
speaker_0NARRATOR
Rumor has it that, if you make it to the final in-person interview, you have to drop your cell phone into a Faraday cage.
speaker_0NARRATOR
This special box completely blocks all phone and radio signals, making sure nobody can secretly record the secret discussions happening inside.
![[YOU] on AI · Ahora](https://particle.news/cdn-cgi/image/format=auto,width=128/https://cdn.particle.pro/url/media/4e91d5b0-2117-58a6-bbc5-99f50507c704/4ba59d61cb1532c3192cbd071151c34c20f282a4cc8dbcff1e4711bdb0859bab)
DanielCORRESPONDENT
Nvidia se encuentra en negociaciones activas para invertir en Perplexity, con una valoración superior a los treinta mil millones de dólares, con una decisión esperada en torno a la presentación de resultados del miércoles.
DanielCORRESPONDENT
El portafolio de Nvidia ya abarca OpenAI, Anthropic, xAI y Safe Superintelligence.
DanielCORRESPONDENT
Sumar a Perplexity extendería su participación desde los chips hasta la capa de búsqueda.
DanielCORRESPONDENT
Perplexity se ha comprometido por separado a utilizar los nuevos CPUs Vera de Nvidia para cargas de trabajo de agentes, estrechando aún más el vínculo entre hardware y software.
JaydenHOST
Mira, Thinking Machines.
JaydenHOST
Ilya, Safe Superintelligence.
JaydenHOST
John, Anthropic.
JaydenHOST
Completely unprecedented situation." Now, all I must ask you is, is it really that unprecedented?
speaker_3HOST
Welcome to Thrilling Threads.
speaker_3HOST
We are pulling apart the incredibly sudden, highly anticipated August launch of Safe Superintelligence, or SSI.
speaker_4HOST
Which is huge.
speaker_3HOST
It's massive.
speaker_4HOST
Scaling up transformer architectures.
speaker_4HOST
We've been throwing more and more compute, more GPUs at the exact same fundamental mechanism.
speaker_4HOST
But SSI, well, they're looking at biological intelligence and realizing that our current algorithms are just they're missing elements.
speaker_4HOST
a foundational component of how living organisms actually acquire and retain knowledge.
Rob KellyHOST
Sam's former head of safety, Dario Amode, runs Anthropic and is battling Sam for what could be the world's largest IPO.
Rob KellyHOST
And two more frontier AI models are being built by former OpenAI folks, Safe Superintelligence from Ilya Sutskovor and Thinking Machines from former CTO Meera Muradi.
Rob KellyHOST
But I want to add something that gets back to Sam himself.
Rob KellyHOST
There's a book Sam recommends to almost everyone he meets, The Beginning of Infinity by physicist David Deutsch.
1 more episode mentions Safe Superintelligence Inc..
Create an account to see the whole feed, search across every transcript, and follow the entities you care about.