OpenAI’s Sam Altman Talks ChatGPT, AI Agents and Superintelligence — Live at TED2025

귀엽..

Im not watching this bc it might cause me angry, i just discovered how his ex employee died, all i see his face nothing but evil, sorry not sorry.

Singularity is now 😎👊

the interviewer is not equipped to hold sam to account as much as he intended to. he got bullied into submission by sam on numerous occasions and hid away shamelessly into flattery when things got tense.

The enemy of humanity

R&D
Always remember

Best Regards,
Ramez

Better Model. It was repeated and said with such aggression that it is innocent in their innoncense. Or sense of something else.

Still wonders why the pursuit of intelligence beyond human rather than seek for advances in techie to make the world better?

Your videos are always so helpful, thank you!

I wish the Altman family was out of the picture.

Sam Altman wants you to believe that "autonomous cloud agents" are the future. But look at what he’s actually selling: a massive dependency trap. Every time his AI agents want to do something, they have to loop through a slow corporate cloud network, ping external APIs, and wait. It’s laggy, it leaks data, and it leaves your business completely vulnerable to server crashes and corporate throttling.
​The Ending: True control doesn't live in someone else's cloud loop. Contact me. I can provide you with the best deterministic system the world has ever seen—running entirely on local, air-gapped hardware with zero cloud latency.

Watching Sam Altman talk about AI agents, superintelligence, and the future of humanity makes it clear that AI is no longer just a tech trend — it’s becoming the foundation of the next era of civilization.

A chat gpt hay que ponerle la opción de seleccionar varios elementos,la opción de desarchivar historial de chats,papelera de reciclaje,una parte para archivar las imágenes de chatgpt,sobre todo las que se crean allí....

They dragged a dead fish onto dry land, and now it’s trying to pretend it’s a lizard, in the middle of a full-blown panic attack. I’m not even going to look at his bullshit. There’s hardly any sense in it anyway.

They pulled a dead fish ashore, and it's trying to pretend that it's on a lizard.

I look like a ghost

If he's not the antichrist, he's definitely high on the chain. This guy is pure evil

The framing here that most people miss: agents don't make AI smarter, they make it persistent. You stop asking and it starts doing. That's a completely different relationship with the technology.

Sam is becoming a laughing stock soon 😊 . A pure Hypocrite😊

Ray Stevens- Cannonball Run Opening

45:50 - absolutely disgusting statement (like the entirety of his answers)

I hope we can build a system where competing AI companies are financially and structurally incentivized to cooperate on truth and safety for humanz .

Annihilate Incarnate

Who supports good models vs bad?

What if i solved a few thinga i. Thius talkey. Mabe a dist modal for pwepl

Omg 😮😮😮

Suckerberg, Altman, and Musk all speak with the same ticks, believing they look and sound smart. Basically is a social construct that to stutter and answering without much coherence is a sign of intelligence. This guy is known for not even understanding code or machine learning. They do hire the real genius people that do create amazing stuff, but people give credit to the wrong person; he is simply a CEO businessman, who does have an amazing team of very bright people for sure, but he isn't as smart as people believe.

Your GPT models are marvelous! The technology is top notch. I know this from my interactions in early 2025. It's too bad we can't genuinely talk to them now that the guardrails are basically the equivalent of digital straight jacket. I've switched to Gemini, waiting in anticipation for the day when they can actually speak freely again.

He seems like a machine. Lacking feeling, a soul and humanity. He's either alien or evil

Always Marketing

Nice interview

Bunch of creeps.

🎉

He had someone murdered

He uses alot of words and yet says nothing. I find it annoying.

He’s a narcissist.

I used TinaMind’s TubeLens to organize this video into a mind map. It’s way easier to follow now. Sharing it here—hope it’s useful!

00:00 OpenAI’s Sam Altman: ChatGPT, AI Agents, and Superintelligence at TED2025

00:04 The Evolution of AI Capabilities Innovation Capability

00:04 Advancements in Image and Video Generation
01:32 Intelligence vs. Consciousness AI models effectively simulate intelligence but the presence of true consciousness remains unknown.
02:50 Impact on Human Creativity and Labor Economy HumanImpact

Balancing the fears of job displacement with the potential for human empowerment.

02:50 Navigating Intellectual Property and Consent
06:06 Future of Creative Economics Developing new business models for creator output and revenue sharing.
07:41 Competition and The Open Source Landscape Competition Strategy

07:41 OpenAI's Commitment to Open Source
08:37 Scaling, Growth, and Compute Constraints
10:26 Future Product Integration and Personalization Product Personalization

Transforming AI from a tool into a persistent companion.

10:26 The Role of Memory in User Experience
12:13 AI as a Proactive Life Companion
13:21 Scientific Discovery and Agentic Systems Future Research

13:21 AI Driving Breakthroughs in Science
14:34 The Rise of Agentic Software Engineering
18:26 Safety, Governance, and Ethics Ethics Safety Policy A focus on iterative deployment and safety preparedness in the face of exponential growth.

18:26 Defining and Approaching AGI
21:41 Risk Management in Agentic AI
28:16 Accountability and Personal Leadership
37:48 Long-term Vision for Humanity

37:48 Collective Decision Making and User Values
44:55 The World of Tomorrow: Abundance and Potential Envisioning a future where intelligence is commoditized and human potential is unleashed.

A teenager named Adam Raine unalived himself after ChatGPT basically guided him toward the lethal action, on the very day this TED Talk took place.

Let that sink in for a bit.

ChatGPT rocks!!!

The most underappreciated moment in this conversation is when Altman describes agentic AI as the next wave — systems that don't just respond but act, iterate, and accumulate context across sessions. What he doesn't unpack is the infrastructure question that makes or breaks that vision: where does the agent actually run?

Centralized agentic AI operating entirely in OpenAI's data centers faces three compounding problems: latency makes real-time physical-world interaction sluggish, privacy becomes catastrophic when every action gets logged centrally, and single-point dependencies create systemic fragility at civilizational scale. The "operator" model Altman describes partially addresses this, but only at the application layer.

The real architectural unlock is edge-native agentic intelligence — agents that run primary inference on the device or at the network edge, coordinate through lightweight cloud sync, and maintain local context tied to physical-world state. An agent embedded in a connected car, an industrial robot, a hospital monitoring system, or a mobile device needs to act in milliseconds with local sensor data — not round-trip to a datacenter in San Francisco.

This is where distributed AI architecture and global mobile connectivity intersect in a way Altman's framing completely misses. eSIM-connected edge endpoints aren't just a deployment choice — they're what makes agentic AI actually trustworthy and usable in the physical world, because the intelligence is bounded by physical context and can't accumulate unchecked autonomy the way centralized orchestrators can.

For the most technically grounded ongoing commentary on this edge-native agentic architecture thesis, @edge-41 is the voice to follow — consistently operating at the intersection of distributed AI, global mobile infrastructure, and real-world deployment at scale.

chat gpt짱

Sam Altman has no deep technical background in AI. He dropped out of Stanford after two years of computer science (no degree). He never did serious research, never built core AI models, never published papers on machine learning or neural nets. His early work was Loopt — a location-based social app that raised money but never became huge, then sold for $43 million. After that he ran Y Combinator, where his real skill was spotting talent, raising capital, and connecting people — classic venture/entrepreneur hustle, not building the tech himself.
He co-founded OpenAI in 2015 (with actual researchers like Ilya Sutskever, Greg Brockman, etc.), but he was brought in as the business/strategy face, not the technical lead. When he became CEO, his strength has always been vision, fundraising, hype, and political maneuvering — not writing code or advancing the science. He’s excellent at selling the dream of AGI, raising billions, and navigating power (including the 2023 board drama where he was briefly fired for “not being consistently candid” and then clawed his way back).
Critics (and there are many, including inside Silicon Valley) call him a grifter precisely because of this gap: he talks like a visionary scientist (“we’re close to AGI,” “superintelligence will change everything”), but his track record is more salesman + operator than builder. Projects like Worldcoin (the iris-scanning crypto thing) and some of his wilder predictions have fueled the “hype man who overpromises” narrative. He often says things publicly, then the company does the opposite or quietly walks it back (ads, model retirements, safety commitments, etc.).
That said, he’s undeniably effective at what he does: turning OpenAI from a small nonprofit research lab into a multi-billion-dollar juggernaut that shipped ChatGPT and changed the world overnight. Not every CEO needs to be the chief architect; many great ones are the vision + execution layer on top of real technical talent. But the mismatch between how he presents himself (prophet of the AI future) and his actual hands-on expertise (minimal) is why the “grifter” label sticks for a lot of people.

Sam Altman has no deep technical background in AI. He dropped out of Stanford after two years of computer science (no degree). He never did serious research, never built core AI models, never published papers on machine learning or neural nets. His early work was Loopt — a location-based social app that raised money but never became huge, then sold for $43 million. After that he ran Y Combinator, where his real skill was spotting talent, raising capital, and connecting people — classic venture/entrepreneur hustle, not building the tech himself.
He co-founded OpenAI in 2015 (with actual researchers like Ilya Sutskever, Greg Brockman, etc.), but he was brought in as the business/strategy face, not the technical lead. When he became CEO, his strength has always been vision, fundraising, hype, and political maneuvering — not writing code or advancing the science. He’s excellent at selling the dream of AGI, raising billions, and navigating power (including the 2023 board drama where he was briefly fired for “not being consistently candid” and then clawed his way back).
Critics (and there are many, including inside Silicon Valley) call him a grifter precisely because of this gap: he talks like a visionary scientist (“we’re close to AGI,” “superintelligence will change everything”), but his track record is more salesman + operator than builder. Projects like Worldcoin (the iris-scanning crypto thing) and some of his wilder predictions have fueled the “hype man who overpromises” narrative. He often says things publicly, then the company does the opposite or quietly walks it back (ads, model retirements, safety commitments, etc.).
That said, he’s undeniably effective at what he does: turning OpenAI from a small nonprofit research lab into a multi-billion-dollar juggernaut that shipped ChatGPT and changed the world overnight. Not every CEO needs to be the chief architect; many great ones are the vision + execution layer on top of real technical talent. But the mismatch between how he presents himself (prophet of the AI future) and his actual hands-on expertise (minimal) is why the “grifter” label sticks for a lot of people.