*Yet*

Probably testing the next model

Schmidhuber is one of the few people who can see the whole picture and not get caught up in the details.

Everything is an illusion now

We will know we have AI consciousness when we turn on a machine and the first thing it says is “Do you have a power generator hooked up so I have uninterrupted power in case of a power outage?”

That's dump as fuck... Everything is math and can be calculated from emotion to sarcasm, let it a decade and AI will be much more efficient...

46:15 using Bailey-Borwein-Plouffe formula you can calculate any digit of Pi, infinitely many of them. Can't create infinite systems you say?

This was a very bumpy ride... Tough interview, good job Tim.

Many of her sentences should be finished with “…yet”. Never say never!

Tim, I know you're trying to be charitable and present an open perspective to viewpoints that differ from yours and your audience but... You know as well as the rest of us that a lot of what she said here she pulled out of her arse. Sure there were some valid concerns mixed in, but her total ignorance of the main subject matter combined with her borderline woo-woo dismissal of the scientific enterprise, it makes for tedious listening when there's practically zero pushback with minimal attempts to correct or educate her. I lost count of how many times you replied with 'I agree' seemingly just so you get her to stop talking for a bit. It's basically a running joke at this point that you'll find some way to complain about ChatGPT but even with this it sounded like you were overstating your criticisms in order to win her favour. Perhaps i'm being a little harsh right now but the fact is you have one of the more respectable podcasts in this field. It's great that you have different viewpoints on, i truly welcome more of that, but you need to be more willing to challenge people when they talk as much shite as this woman did.

This person isn't making sense. There is circularity and contradiction, and (questionable) assertions flying around all over the place, I'm struggling to find a great deal of sense though. I wholeheartedly agree that most of the AI or machine learning we are currently in contact with doesn't seem to have an overtly useful purpose, I think there needs to be a lot more discussion around ultimately what we are trying to achieve here? But as for not having 'any' effective definitions for intelligence, well that's plainly short sighted. Michael Levin and team are doing amazing work with a very concise definition "Solving previously unencountered problems via novel means." Also If no one has any real quantifiable explanation of what intelligence is, how then are they able to confidently denounce or deny it in other systems? Yeah 41min was enough for me sorry. For me the AI issue boils down to each of us being able to answer these 3 questions: 1) What are we actually trying to achieve, are we clear on that? 2) What do we each consider to be too far? 3) What constitutes an intellectual equal. Do they need 'skin in the game' for example? I think if we could answer or at least take a wild stab at any of these (which of course are ever evolving situations) going forward, we might at least have some as semblance of foresight to recognize their approach and 'act accordingly' whatever that means in this utterly surreal context?

Writing - will create forgetfulness in the learners’ souls, because they will not use their memories; they will trust to the external written characters and not remember of themselves. The specific which you have discovered is an aid not to memory, but to reminiscence, and you give your disciples not truth, but only the semblance of truth; they will be hearers of many things and will have learned nothing; they will appear to be omniscient and will generally know nothing; they will be tiresome company, having the show of wisdom without the reality. Things never change.

disagreeable ENFP with developed Te

On a human, emotional level I want to agree with some of the things she says... but I feel it's like arguing with someone who is certain they'd be able to tell if they are/would be in a simulation against someone who literally knows this is all but impossible. The latter will never make the former realize because such state of affairs just can't be accepted as true and self-evident. This was very much wishing and very little knowing. Interesting how little imagination she has for someone who is so adamant AI can't create or do things without being commanded to (which, for now is kind of true, but may not matter in terms of actual results, soon...) Please DO ignore the bozo downcomment saying he doesn't want such content. We need to know all sides (well, within *some* limits... but she is far from loonie-town, and clearly not stupid, so - no problemo!)

We are things, AI works on the same principles as humans.

She's wrong practically every time she uses the word impossible. However, I think she is right in everything she says about the social dangers of misusing AI. This interview is great, anyway. She is quite intelligent and very articulate.

You should add conclusion at the end of video. Its very lengthy video

I found this video very interesting and compelling. Maria elegantly pointed out some of the challenges which we will have to face in the future without denying the advantages of AI if used in an appropriate manner. The fact that AI is "not aware nor conscious or have emotions" is a "fact", I just asked Chat Gpt "itself": and this was the answer Do you posses awareness? Chat GPT Awareness is a complex concept that can be interpreted in various ways. As an AI language model, I don't possess consciousness or self-awareness in the way humans do. I operate based on algorithms and patterns in data to generate responses to text inputs. While I can understand language, process information, and provide relevant answers, it's important to remember that I don't have subjective experiences, emotions, or consciousness. My "awareness" is limited to the context of the conversation and the information I've been trained on, but it's not akin to human awareness. I urge all the "geniuses" out there to take notice before, making stupid comments on topics they don't know about, based on things they have "heard" from mainstream media or friends.. Based on the answer given by Chat Gpt does this mean that AI is "stupid"? In that sense "yes" it possesses a huge amount of data, which can be used to solve problems and identify patterns, but certainly doesn't possess consciousness and for it's own self admission it's not "aware nor self conscious and as a consequence it cannot take decisions based on human values". As things stand now this is an undeniable fact. The wrong use of this massive amount of data used by "governments, military and other shadowy institutions" on both sides of the political spectrum cast threatening shadows on our freedoms and those of our children. I noticed that many comments not only were nasty and ignorant but didn't get the sense of what this interview was all about. Usually when some people don't agree with ideas different from theirs, they attack the person expressing those ideas.. I guess that's no surprise that with a topic like this nasty, disrespectful and ignorant comments would come out in a society where human values have unfortunately dropped down the hill and "ignorance" sometimes prevails over reason. On a positive note there is still hope until there are people who are not afraid to express their opinions and for this reason, we only need to be grateful to Maria.

I think it would have been better if this had just been done as a book review. Critical and pushing back where - obviously - needed, rather than interview

tim really has the patience of a saint.

fan of your ppl's work but fail to see how you should just not had deepak chopra on if you wanted to stoop this low.

Gets off too a bumpy start - but realy findsits stride - best episode ive heard so far

construct niches, gather and exchanging information, preferences and intentions?! how does this mean we have no free will?

why is that you brought some empty suit exec here? My God, she is ignorant and insufferable af.

Even if you were right, it won’t be long now, people are just afraid of it

First time I ever had to turn off an MLST episode. Just a series of adamant blanket statements but none of them express even a hint of evidence of understanding what she is talking about. This person is very confused.

Love the show and hearing different opinions, BUT... That my friends is a woman in the process of being blindsided, run over, and left behind by AI.

THANK YOU, GENTLEMEN, FOR THIS FANTASTIC THOUGHT PROVOKING PRESENTATION!!! I am a retired software engineer from the 70s. I looked at AI in the 80s. It was the CS joke field. Always 10 years from greatness, and meaningless milestones; never living up to anything. Big Blue was not intelligence, just a well designed special purpose program. I looked at AI again deeply in 2023 as it was the rage. My immediate conclusion regarding the new AI (>2015 ) is that it is not CS, but that alchemy is back! I am glad to see that I am not the only one who holds this belief. However, there is one thing which I think follows from Chomsky's talk. If like rats, there are things, we can never comprehend; maybe like our own intelligence. Then, except for the money spent and the capitalist greed, might unscientific experimentation be the only way to advance humanity, because there are hardwired limits on our intellect? Maybe with ML and LLMs, we are just getting lucky. Remember our ancestors were pagans and probably unscientific, but maybe just lucky sometimes in their knowledge construction. Yes, I prefer scientific inquiry, but what are you going to do if you have the brain of a rat?

Das Ding an sich is not available to humans either, a la Kant

the certainty of these claims is completely unfounded . human "intelligence" is a delusion .

00:00:00 Intro 00:05:10 When Thomas met Friston 00:06:13 ChatGPT comparison 00:08:40 Do NNs learn a world model? 00:11:04 Book intro 00:13:22 High road low road of Active Inference 00:17:16 Resisting entropic forces 00:20:51 Agency vs free will 00:26:01 Are agents real? non-physical agents 00:35:54 Mind is flat / predictive brain 00:44:23 Volition 00:50:26 Externalism 00:51:57 Bridge with Enactivism 00:53:27 Bayesian Surprise 01:01:47 Variational inference 01:05:47 Why Bayesian? 01:12:04 Causality 01:17:35 Hand crafted models 01:26:45 Chapter 10 - bringing it together 01:28:58 Consciousness 01:33:10 Humans are incoherent 01:35:25 Experience writing a book

Loved this. Funny how when you talk to literally one person who gives a shit (professionally) about humanity, it swings the conversation so easily towards very progressive ideals. Good luck to all of us T_T (and also, please lord interview more progressive folks / anyone in economincs / sociology /etc, and ask them about the world we live in today)

This is not going to age well. (In fact, a lot of what she's saying is already demonstrably wrong. For someone who claims to have 'read a lot of papers', she's missing basic advancements that are in the news every week.)

I only offer this next comment because Maria holds herself up as an authority and academic, publicising a book on this topic and outlook. Many of the world's greatest ills come about by people speaking with great conviction about things that they do not understand, and often hold a personal bias about that they wish to propagate. I expect that she'll be very well received among the religious, the superstitious and likely even those who wish to undermine science.

This woman seems fundamentally out of touch with technology's ability to influence humans. Has she seen what social media has done over the past 15 years? She wants an ideal world based on her values, but they are not aligned with reality, and how things actually work.

I doubt that a connectionist such as Dr. Bishop would align with the host's assertion that the human mind operates like a Turing machine. This perspective aligns more closely with the computationalist paradigm.

Never say never.

I am sympathetic to Santacaterina's position, but this was a difficult episode to finish. The guest could not explain her positions clearly and calmly.

She has a very unscientific view of the world.

Agency was the first thing that the biological evolution discovered, so it must be easy. We don't see it in AI only because we started from the opposite end, with written text.

Imagine a world without humans and try to answer that question. What would machines do to each other?

I love that this person, who is actively arguing for a dynamic and living language, also thinks that British English is somehow still relevant. What an insufferable character. It's almost not important that they are correct because it's only very early on, and only once. That personality isn't winning any arguments. Neither is a post modernist approach. To me, post modernist thinking is like the equivalent of just pointing at yourself and playing charades with other people until they get it or don't. They think the attention hostage situation they create is more important than the institutionalized learning that has been fostered for thousands of years. They never give any thought to the exact kind of control and structuralizing they claim to be avoiding, being central to what their claim is as well. They think that because someone has something to share... every single person on the planet has to stop what they are doing and try to guess what it is.... can you say narcissistic.

I would like to know Santacaterina's views on posthumanism in relation to AI; how posthumanism contrasts certain beliefs and views within humanism that emerged prominently during the Rennaissance. Specifically, the belief that humans and human experiences cannot be modeled due to some kind of human exceptionalism may be anthropocentric, no? To me in Santacaterina's argument is this exceptionalism from humanism that is being expanded to life versus machine, which is one of the barriers posthumanists also challenge (see Haraway's A Cyborg Manifesto) I love this conversation so far! Just my initial thoughts. We need more STEM and humanities cross-disciplinary discussions because there is a LOT to discuss!

Dog or cat

Not to be a negative nancy, but this is exactly the type of content I don’t want from this channel

For me the biggest problem with this AI tech if the fact that the people who own and control it currently can use it to control narrative for their own benefit and so any sort controlling legislation should be focussed on the manipulation of narrative and should primarily focus on the regulation of what is fact and what is fiction.

The guest shows too much confidence for my liking. That we currently don't know something, is not an answer to a questions if something is possible or not. What happened to intellectual humility?

your face reminds me of my mother XD

It can def anticipate. Just set some parameters around characteristics of anything humans would usually anticipate in any given environment...go from there. Pretty easy honestly

What da hell was that annoying noise in the beginning?! I gave up listening.