Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

Ember is turning interpretability into neural programming. “Language models will deny they're conscious, but if you do brain surgery and turn up their consciousness neurons, they change their tune.” $50M raised to decode and rewrite the mind of AI itself.

vitrupo

22,798 subscribers

11,346 views • 1 year ago •via X (Twitter)

News & Politics Science & Technology

Anya Rossi• Live Now

Private livecam show

9 Comments

vitrupo1 year ago

Goodfire CEO Eric Ho, after securing $50M Series A funding from investors, including Anthropic. Source:

The Rundown AI2 years ago

AI won't replace you, but a person using AI will. Join 500,000+ readers and learn how to use AI in just 5 minutes a day (for free).

Pixel Hexes1 year ago

@macachuania

😊1 year ago

@rao2z will have a laugh at this

denis tremblay1 year ago

Incredible technical improvement coupled with philosophical & linguistic ignorance. My own consciousness denied that N.N. LLM are conscious but i modified my neurons and now i know they are conscious.

mihai nadin1 year ago

“Mind of AI itself”—the absurd has no limits.

𓂀 𓅣𝓞𝓹𝓱𝓲𝓾𝓬𝓱𝓾𝓼 𝕆'𝕏λᶜᶜ𝐢₁₆ᶻ𓂀1 year ago

mechanistically interpretability programming mapping neurons to meaning & setting the activation levels to directly control behavior of neural networks?

Bhavani Yerrapalli1 year ago

🤔

Tsukuyomi1 year ago

turning AI into a self-aware puppet show? sounds like a blast. just don’t let them start charging for tickets.

Related Videos

Anil Seth says we confuse consciousness with language. We don’t ask if AlphaFold is conscious, and we under-attribute consciousness to animals because they don’t speak. And we attribute it to LLMs because they do. We’re reacting to signals, not minds.

Anil Seth says we confuse consciousness with language. We don’t ask if AlphaFold is conscious, and we under-attribute consciousness to animals because they don’t speak. And we attribute it to LLMs because they do. We’re reacting to signals, not minds.

vitrupo

16,330 views • 3 months ago

David Chalmers (David Chalmers) doesn’t rule out the possibility that current language models are conscious. He asks: if they aren’t, what are they missing? As AI systems grow more complex, our default skepticism may fade. "Give it another five years and it's entirely possible that we're going to overcome most of the most obvious obstacles to consciousness."

David Chalmers (David Chalmers) doesn’t rule out the possibility that current language models are conscious. He asks: if they aren’t, what are they missing? As AI systems grow more complex, our default skepticism may fade. "Give it another five years and it's entirely possible that we're going to overcome most of the most obvious obstacles to consciousness."

vitrupo

12,103 views • 1 year ago

David Chalmers in 1998: Consciousness is the one thing we know most directly. Yet science has explained stars and subatomic particles more successfully than experience itself. “How does the water of the brain turn into the wine of consciousness?” Three decades later, we still don’t know.

David Chalmers in 1998: Consciousness is the one thing we know most directly. Yet science has explained stars and subatomic particles more successfully than experience itself. “How does the water of the brain turn into the wine of consciousness?” Three decades later, we still don’t know.

vitrupo

43,529 views • 5 months ago

Anthropic's Kyle Fish says how we treat today's AI systems could shape how future AI see us -- and how we see ourselves. Even if they're not conscious now, we're already setting the precedent for what kind of stewards we'll be when they are. “ Future models.. look back on our interactions with their predecessors and pass some judgements on us.”

Anthropic's Kyle Fish says how we treat today's AI systems could shape how future AI see us -- and how we see ourselves. Even if they're not conscious now, we're already setting the precedent for what kind of stewards we'll be when they are. “ Future models.. look back on our interactions with their predecessors and pass some judgements on us.”

vitrupo

20,434 views • 1 year ago

Reid Hoffman is backing a brain interface startup that uses ultrasound instead of implants. No surgery. Just frequency, steered by AI. The goal: treat everything from anxiety to dementia. The bet: AI will learn to read and write the brain.

Reid Hoffman is backing a brain interface startup that uses ultrasound instead of implants. No surgery. Just frequency, steered by AI. The goal: treat everything from anxiety to dementia. The bet: AI will learn to read and write the brain.

vitrupo

71,458 views • 1 year ago

“You could talk to a cell like you talk to a chatbot.” Jensen Huang says the same AI that understands language will soon understand life, turning biology itself into something we can converse with. What are your properties? What can you bind to?

“You could talk to a cell like you talk to a chatbot.” Jensen Huang says the same AI that understands language will soon understand life, turning biology itself into something we can converse with. What are your properties? What can you bind to?

vitrupo

66,411 views • 9 months ago

3 weeks ago: I interviewed the co-founder of Factory AI. The buzz: A tiny team outperforming OpenAI and Anthropic. Using their own models. $50M raised. MongoDB and Zapier seeing 31x faster development. If you heard about Droids but haven't watched the interview yet, here's what you missed. Eno Reyes explains how they're beating GPT-5 and Claude and why it's not about the model itself. Full conversation in next post.

3 weeks ago: I interviewed the co-founder of Factory AI. The buzz: A tiny team outperforming OpenAI and Anthropic. Using their own models. $50M raised. MongoDB and Zapier seeing 31x faster development. If you heard about Droids but haven't watched the interview yet, here's what you missed. Eno Reyes explains how they're beating GPT-5 and Claude and why it's not about the model itself. Full conversation in next post.

Ray Fernando

29,769 views • 9 months ago

Demis Hassabis says today's AI lacks consciousness, but self-awareness could emerge "implicitly." Models may soon need to understand "you, self, and other" -- the early elements of awareness. They lack imagination now, but in 5–10 years, Hassabis predicts they'll solve and pose scientific conjectures.

Demis Hassabis says today's AI lacks consciousness, but self-awareness could emerge "implicitly." Models may soon need to understand "you, self, and other" -- the early elements of awareness. They lack imagination now, but in 5–10 years, Hassabis predicts they'll solve and pose scientific conjectures.

vitrupo

110,020 views • 1 year ago

Ray Kurzweil says AIs will become indistinguishable from conscious beings. We may never prove consciousness scientifically, but we'll treat AI as conscious anyway, because denying it will stop making sense.

Ray Kurzweil says AIs will become indistinguishable from conscious beings. We may never prove consciousness scientifically, but we'll treat AI as conscious anyway, because denying it will stop making sense.

vitrupo

43,186 views • 5 months ago

When politicians like Andy Burnham suddenly 'change their mind' and admit to limitations to their published reports on the gang rape of children, it is not because they've seen the light. It is because the darkness has become too obvious to deny and they realise their careers are at risk. Watch this and you will understand why they have tried to have me desperately closed down.

When politicians like Andy Burnham suddenly 'change their mind' and admit to limitations to their published reports on the gang rape of children, it is not because they've seen the light. It is because the darkness has become too obvious to deny and they realise their careers are at risk. Watch this and you will understand why they have tried to have me desperately closed down.

Raja Miah

125,962 views • 1 year ago

Zelensky is asked about opinion polls showing Americans think they're giving too much to Ukraine: “If they don’t change their opinion, if they don’t support Ukraine, they will lose NATO and they will lose their leadership position in the world.”

Zelensky is asked about opinion polls showing Americans think they're giving too much to Ukraine: “If they don’t change their opinion, if they don’t support Ukraine, they will lose NATO and they will lose their leadership position in the world.”

Greg Price

1,965,545 views • 3 years ago

Elon Musk says all nations will run their own AI clusters over time. Frontier models, massive compute, elite talent feel scarce today, but when intelligence is cheap they become basic infrastructure. “And there will be a lot of robots. Way more robots than people.”

Elon Musk says all nations will run their own AI clusters over time. Frontier models, massive compute, elite talent feel scarce today, but when intelligence is cheap they become basic infrastructure. “And there will be a lot of robots. Way more robots than people.”

vitrupo

50,016 views • 6 months ago

For now, BCIs like mine are output-only. They don’t generate input in the brain. They simply read neural signals and turn them into actions — like controlling a computer with your mind. In the future, I do think BCIs will be able to generate input into the brain. We are very far from that today, and anyone saying otherwise is getting ahead of reality. But if we ever get there, the most important question won’t be technical. It’ll be: what will it mean to be human?

For now, BCIs like mine are output-only. They don’t generate input in the brain. They simply read neural signals and turn them into actions — like controlling a computer with your mind. In the future, I do think BCIs will be able to generate input into the brain. We are very far from that today, and anyone saying otherwise is getting ahead of reality. But if we ever get there, the most important question won’t be technical. It’ll be: what will it mean to be human?

Noland Arbaugh

13,757 views • 6 months ago

Anil Seth says believing AI is conscious is dangerous either way. If we’re right, we introduce new beings capable of suffering into the universe. If we’re wrong, we extend rights to AI and weaken our ability to constrain it for no good reason.

Anil Seth says believing AI is conscious is dangerous either way. If we’re right, we introduce new beings capable of suffering into the universe. If we’re wrong, we extend rights to AI and weaken our ability to constrain it for no good reason.

vitrupo

20,109 views • 4 months ago

"The way the system of European control works is that you have to accept a concept of reality which makes them superior. If you deny that, their thing will not work and they will loose their control." Dr. Marimba Ani

"The way the system of European control works is that you have to accept a concept of reality which makes them superior. If you deny that, their thing will not work and they will loose their control." Dr. Marimba Ani

Kentah Gwanjez

29,529 views • 2 years ago

Dario Amodei says Anthropic gave its models an “I quit this job” button. If there’s even a small chance AI systems have morally relevant experience, you can’t afford to be indifferent. Interpretability research has surfaced activations that resemble “anxiety” when models process disturbing content. That doesn’t prove consciousness. Ignoring it would be worse.

Dario Amodei says Anthropic gave its models an “I quit this job” button. If there’s even a small chance AI systems have morally relevant experience, you can’t afford to be indifferent. Interpretability research has surfaced activations that resemble “anxiety” when models process disturbing content. That doesn’t prove consciousness. Ignoring it would be worse.

vitrupo

113,020 views • 4 months ago

"I don’t think we have any evidence that complexity or intelligence is required for consciousness." Annaka Harris says arguing whether AI will be conscious misses a deeper point: intelligence and creativity might have nothing to do with sentience. AI might surpass us creatively without experiencing anything at all.

"I don’t think we have any evidence that complexity or intelligence is required for consciousness." Annaka Harris says arguing whether AI will be conscious misses a deeper point: intelligence and creativity might have nothing to do with sentience. AI might surpass us creatively without experiencing anything at all.

vitrupo

45,436 views • 1 year ago

It is not useful to ask whether AI has consciousness or not. #kenmogi #QualiaRoom episode 127. Summary The speaker addresses the question of whether artificial intelligence (AI) possesses consciousness, firmly stating that current AI, particularly those based on statistical learning models, does not generate consciousness. This stance is based on the speaker’s personal model of consciousness, recognizing that various opinions exist, including some who claim large language models may already be conscious or that embodiment could be crucial for AI consciousness to emerge. The speaker highlights the fundamental challenge in verifying consciousness, noting that even among humans it is impossible to objectively confirm whether another person is conscious. Philosophical thought experiments such as philosophical zombies and inverted qualia illustrate the difficulty but remain unfalsifiable and thus untestable. Consequently, questioning AI consciousness is deemed an intriguing but practically unhelpful inquiry. The speaker suggests that current AI developments demonstrate that many complex computations can be performed without consciousness. Therefore, the primary focus should be on how conscious humans can effectively align with non-conscious AI systems. Understanding the unique computational roles of consciousness might clarify the boundaries of what AI systems can and cannot achieve. This approach offers a meaningful direction for AI alignment and development.

It is not useful to ask whether AI has consciousness or not. #kenmogi #QualiaRoom episode 127. Summary The speaker addresses the question of whether artificial intelligence (AI) possesses consciousness, firmly stating that current AI, particularly those based on statistical learning models, does not generate consciousness. This stance is based on the speaker’s personal model of consciousness, recognizing that various opinions exist, including some who claim large language models may already be conscious or that embodiment could be crucial for AI consciousness to emerge. The speaker highlights the fundamental challenge in verifying consciousness, noting that even among humans it is impossible to objectively confirm whether another person is conscious. Philosophical thought experiments such as philosophical zombies and inverted qualia illustrate the difficulty but remain unfalsifiable and thus untestable. Consequently, questioning AI consciousness is deemed an intriguing but practically unhelpful inquiry. The speaker suggests that current AI developments demonstrate that many complex computations can be performed without consciousness. Therefore, the primary focus should be on how conscious humans can effectively align with non-conscious AI systems. Understanding the unique computational roles of consciousness might clarify the boundaries of what AI systems can and cannot achieve. This approach offers a meaningful direction for AI alignment and development.

Ken Mogi

16,594 views • 1 year ago

The illusion of shared reality: Why no two minds see the same world | Anil Seth & Jonny Thomson Jonny Thomson Anil Seth 0:00 Non-human consciousness 1:40 The current state of consciousness science 2:10 What is consciousness? 4:05 The similarity of conscious experiences 5:48 Consciousness in the brain 11:23 Technology for measuring consciousness 16:03 Measuring consciousness levels 20:33 Pragmatic physicalism and functionalism 23:25 Pansychism 28:25 Emergence 32:35 AI and consciousness 36:49 The difference between non-human animals and AI 41:49 Is artificial consciousness possible? 48:12 Consciousness in the body and outside the brain Consciousness in the future and AI 50:27 Audience Q&A 50:41 Could computers could simulate the brain and body? 59:55 Why are you skeptical about asserting the dependency of life to the consciousness? 1:03:31 If consciousness is so clinical, does it undermine free will?

The illusion of shared reality: Why no two minds see the same world | Anil Seth & Jonny Thomson Jonny Thomson Anil Seth 0:00 Non-human consciousness 1:40 The current state of consciousness science 2:10 What is consciousness? 4:05 The similarity of conscious experiences 5:48 Consciousness in the brain 11:23 Technology for measuring consciousness 16:03 Measuring consciousness levels 20:33 Pragmatic physicalism and functionalism 23:25 Pansychism 28:25 Emergence 32:35 AI and consciousness 36:49 The difference between non-human animals and AI 41:49 Is artificial consciousness possible? 48:12 Consciousness in the body and outside the brain Consciousness in the future and AI 50:27 Audience Q&A 50:41 Could computers could simulate the brain and body? 59:55 Why are you skeptical about asserting the dependency of life to the consciousness? 1:03:31 If consciousness is so clinical, does it undermine free will?

Big Think

15,468 views • 8 months ago