- Видео 1
- Просмотров 13 246
mriz
США
Добавлен 15 мар 2018
what day is 33 years ago?
Synthetic Sentience: Can Artificial Intelligence become conscious? | Joscha Bach | CCC #37c3
from official page:
media.ccc.de/v/37c3-12167-synthetic_sentience
Despite the rapid progress of AI capabilities, the core question of Artificial Intelligence seems to be still unanswered: What does it take to create a mind? Let us explore the boundaries of AI: sentience, self awareness, and the possibility of machine consciousness.
After many attempts to build AI models that are smarter than human beings, we find ourselves confronted with a family of surprisingly successful systems that match many of our abilities through text prediction and text/image correlation. The limits of these approaches are presently unclear, and while they work in very different ways than our minds, they pose the q...
media.ccc.de/v/37c3-12167-synthetic_sentience
Despite the rapid progress of AI capabilities, the core question of Artificial Intelligence seems to be still unanswered: What does it take to create a mind? Let us explore the boundaries of AI: sentience, self awareness, and the possibility of machine consciousness.
After many attempts to build AI models that are smarter than human beings, we find ourselves confronted with a family of surprisingly successful systems that match many of our abilities through text prediction and text/image correlation. The limits of these approaches are presently unclear, and while they work in very different ways than our minds, they pose the q...
Просмотров: 13 251
I'm missing a distinction between consciousness and the contents of consciousness. To me it's an intuitive idea that the contents of consciousness are just virtual representations, what isn't intuitive to me is how consciousness itself arises in a mechanical universe.
the way i understand it is, consciousness is the function (process) that integrate data from all sensory apparatus into coherent now. so, the consciousness itself it not an object that you can put your finger on, but the process of how your appearance stable in nowness.
stylistic horror. Shirt, cheap old sofas, name and logo of conference
24:51 gods were a solution to organize society at scale (synchronized across people with rituals and temples, competing for space with our Selves in our own heads)
37:30 I'm not an optimist with respect to AGI, but I'm not a pessimist. I am an expectationalist, it's going to happen and we are not prepared
23:00 a God is a self that spans multiple minds
14:30 neurons create consciousness in order to see what it would be like if there were a person that cared
11:45 The conductor of the symphony is another instrument, tasked with listening to a few instruments at a time and increasing their harmony. I find a parallel between this and the IFS Self
20:00 The development of consciousness
13:00 consciousness is not intelligence, sentience, agency, self,
10:30 consciousness is second order perception, the perception that I am perceiving things
I'm struggling with consciousness being virtual, and as Joscha adds, representational. What is Joshcha's position on whether consciousness is real or not? Consider the quale, red. I accept that red is representational. It is not a property of any subject of our perception. Red is only conjured up in our minds. I expect that Joscha would characterize red as virtual. Does he then mean that red is not real? My belief is that red is real. "Red" is massless. It is not pigment. My best guess is that it results from a bioelectric release of energy that evolution has figured out how to render as an impression of consciousness. In other words, red has a material existence, where material refers to matter and energy, between which Professor Einstein tells us there is an equivalence. If this is so, it gives us the final physical correlate of consciousness. It would not solve the "hard problem." But, it would narrow the search. Returning to Joscha's comments, there appears to be a reasonable argument for a material existence of consciousness. I am unable to sort out if Joscha would agree. The words virtual and representation do not resolve his position on the question to me.
I think there has to be a distinction between consciousness (the screen) and the content of consciousness (the movie that's projected onto the screen). It's intuitive to me how the content of consciousness is just virtual and representational. What I don't get is how the phenomenon of consciousness itself suddenly emerges in a purely mechanical universe.
@@shekhinah5985 Where I fall off the log is just the opposite. To me the cropper is the word "virtual." By virtual, I take it that Joscha refers to something that comprises neither matter nor energy: an intangible. Yet, this virtual entity affects behavior. This requires an interaction with the body, something with a physical instantiation. At this point, we run into the interaction problem, which can be summarized in Newton's equation of motion F = ma. Something that is virtual has no mass or energy and cannot exert a force on something physical. The virtual claim immediately implies a Cartesian duality as a consequence of Joscha's argument. Moreover, displaying "dancing figures" on the "screen" of consciousness requires the expenditure of energy, at least.
Excellent speaker; ahead of his time, making an important landmark which deserves respect now, as he will no doubt be remembered for it in the future
I love JB but maybe AI did his clothes? 😂
Nooo. We cant even explain 1 consciouss experience we have. What AI are you dreaming????
I was so pleased to hear Joscha comment on Julian Jaynes... it's something I always wanted him to comment on and here he finally has! I laughed out loud when he said Jaynes chose the wrong title for his book and should've called it, "The Origin of the Personal Self in the Breakdown of the Polytheist Mind"... so good, and probably right to say so!
Bach is such a phenomenon. The final passage of his speech is simply fantastic. I wouldn't be surprised if this is going to be watched in the future, after everything has expanded out from the singularity.
JB is generations ahead - from a different galaxy. He is the only scientist I sm aware of who is integrating multople disciplines especially philosophy and theology and even scriptures - into current science and A1. Beautiful and amazing. ❤
Quite a few determined and sincere scientists during lifes work in research do end up in confidence of existence of God.
Beautiful
Thank you for the talk. I believe there are ideas and threads that you are missing that may illuminate a number of answers to the open questions in your talk. I have been working on engineering solutions to consciousness since 2018 when there was a quantim leap in the understanding of AI. If interested, I would be glad to offer my email to begin. Steve T.
10:18
Why do u reupload the original video from youtube? ruclips.net/video/cs9Ls0m5QVE/видео.html
Oh that sucks and we are all here in the comments
Are you sure? The link you point to seems to have been uploaded more recently than this one...
@@deanwbrandon I recall it was missing from the regular media CCC feed and I was watching it on the website. This upload here probably was first.
He is 50-100 years ahead. How do I make money out of fhat insight now? 😅
exactly same thoughts. everything is obvious to him apparently
brilliant stuff. The question is how do we actually, right now, this minute, find a way to start moving an essential part of ones self into the new digital substrate.
Nanotechnology, specifically synthetic neurons/neural prostheses… the idea being that each synthetic neuron binds to a living one and then essentially emulates/clones its behavior, until the biological neuron becomes obsolete (cell death) and the synthetic one continues functioning as its former organic counterpart; this leads to the philosophical question of what precisely would happen to the subjective “ID” or ‘self’
@@mettattem nice idea but this has to be at least 20-30 years out. I want to get started uploading right now.
@@travisporco break the illusion of time?
@@mettattemexcellent comment thanks
Hat der sich sein T-Shirt auch selbst generiert auf ein DMT Trip 😅
not sure if the subtitles were accurate but "meat space" is my new favourite term for "in real life"
That's the source. William Gibson's Neuromancer
Absolute bs. trying to recreate the mind without understanding where consciousness arises is a farce. Programmers think they can somehow program consciousness or that it somehow rises out 1's and 0's is grossly misunderstanding the nature of our reality. Trying to model the brian wont get you to consciousness either, for the brain isnt where consciousness arises, the whole body is conscious. All digital computers can be boiled down to logic gates, at its base layer consciousness cannot be boiled down to logic gates, neurons are also not where consciousness arises, nor are microtubules. The whole section of 'what is consciousness' is incorrect too. Maybe try listening to philosophers like bernardo kastrup, clinical psychologists like donald hoffman, even sam harris has a better grasp on consciousness than this talk did. So many misconceptions parroted in this talk, its mind boggling. It's the hubris of programmers that think they can do what took biology 4 billion years to do. Now I do agree AI will approach a place where it may mimic something that resembles consciousness just due to computing and memory, but lets not confuse this mimicry of consciousness with real consciousness.
He also mentions that the mental universe by consciousness was an idea put forth by genesis 1, but this idea is literally millenia older proposed by the vedic hindus in 6000 BC or older, well pre-dating 'genesis 1', if this lecture is to be taken seriously, then a lot more research into the subject of consciousness is needed by this presenter, hes a few millennia behind on the mind.
Like the talk but Love the shirt!
.. and if we're lucky, she'll take us with her and integrate us with her into a global planetary agent... göttliche Gedanken
Utility monster in our society are children and pets! 😊
Would be nice to see some ideas from Researchgate article 365369026_Theory_of_Consciousness been incorporated. Maybe, some of them are useful here.
The question is who cares? Humans will be in 200 years gone.
I had to turn it to 0.75x speed to follow his talk. Still cannot say that I fully comprehend it. Joscha is simply on a higher level. It seems all these ideas and concepts are super intuitive to him while I am busy turning his sentences into sth that my monkey brain can make sense of. 😅
Heck yea! Keep rewatching it, you will pick up another Gem or two every time 🦬🙏
same for me, but the fact that you are here and wrestling with this hard ideas is worthwhile
Oddly enough, I find the content most enjoyable and easy to absorb while watching at 125% speed 🙃
I just do the pause and seek back
@@mettattem, now we're talking! 😏 That's how it should be
i'd watch this a million times
Agreed! 🦬🙏
what would be the AI analog of hormones/hormonal signaling?
Nothing yet, think about what he said about sentience vs. consciousness. They might be conscious but there is no sentience yet. When we get there (or if someone is already simulating it), that might be the place. But ChatGPT does not have it today.
E.g. hormones are changing the weighting of parameters. Of course this could be done today already, but hormones are reacting on circumstances. The models today don't know circumstances yet. How this would happen in detail, how could these mechanisms look like, might be one of the many interesting different fields to research on.
Certainly! Here's a consolidated list of virtual factors in AI, simulating aspects of biological processes: 1. **Adaptation Factor (Learning Rate Adjustment):** - *Mechanism:* Monitors task changes or input complexity. - *Implementation:* Dynamically adjusts the learning rate to adapt to new challenges or prioritize stability. 2. **Focus Hormone (Attention Modulation):** - *Mechanism:* Monitors task importance and resource allocation. - *Implementation:* Modifies attention mechanisms to enhance performance in critical areas. 3. **Memory Reinforcement Factor (Memory Consolidation Signal):** - *Mechanism:* Evaluates the significance of learned experiences. - *Implementation:* Enhances weight updates for specific memories, reinforcing important knowledge retention. 4. **Resilience Hormone (Adversarial Defense):** - *Mechanism:* Detects adversarial attacks or unexpected input variations. - *Implementation:* Triggers defensive mechanisms to resist adversarial manipulations. 5. **Curiosity Hormone (Exploration-Innovation):** - *Mechanism:* Monitors familiarity with the environment or task space. - *Implementation:* Adjusts exploration-exploitation trade-offs, fostering the discovery of new strategies. 6. **Stress Regulator:** - *Mechanism:* Monitors workload or error rate. - *Implementation:* Dynamically adjusts stress levels, influencing the model's response to challenges. 7. **Collaboration Signal:** - *Mechanism:* Detects opportunities for cooperation or knowledge sharing. - *Implementation:* Triggers a collaboration signal, encouraging information exchange among AI instances. 8. **Elasticity Factor (Resource Allocation):** - *Mechanism:* Observes computational resource availability. - *Implementation:* Modifies resource allocation to scale computational intensity based on demand. 9. **Energy Conservation Signal:** - *Mechanism:* Monitors power consumption or environmental factors. - *Implementation:* Activates an energy-saving signal, prompting the AI to prioritize efficiency during resource scarcity. 10. **Emotional Response Simulation:** - *Mechanism:* Analyzes user feedback sentiment or context. - *Implementation:* Simulates emotional responses in AI, affecting tone or expressions for improved user experience.
@@gerdhuebner2848 This is an incredible and insightful response to a notion of curiosity I had while listening. That's a great connection on hormones can functionally be reduced to weight modifiers. I love the list, going through them one by one
Thank you, Joscha! Just 2 weeks ago I thought about what would I ask him, if I had a chance for one question. My question would have been: What does Michael Levin's Planaria/Barium experiment tell to you?🎻🤩
He’d say: “nevermind that shit, Im busy!”
You are wright, he might say he is busy. But you are wrong about the matter itself...listen again around 32-35 min...and in the end where he mentions Mike Levin again.
The other one is 58.40...allthoug I assume the group Joscha mentions there is inspired by another study of Mike Levin, the study on xenobots.
I trust you've found the couple of podcasts they have been on together. Bach says he expects AGI eventually - so would he expect that AGI to combine itself with Levin's xenobots infused with planeria's "immortality"? AGI might be an eventuality, but I'm not so certain it will come about before the species destroys itself. I'm very pessimistic about the human species ability to survive too much further into the future beyond small subsistence groups after a/some massive global societal collapse(s) if not total extinction. Levin's work has paradoxically struck me as possibly one of the nails in the coffin of humanity but conversely also as a glimmer of hope in that even without humans consciousness will rise again in another, possibly "better", form that could manage to succeed where humanity has failed. Bach + Levin = the two most interesting people on the planet.
luckily for you bach has actually spoken with levin at least once (search their names plus "theories of everything" which is the podcast they were co-guests on with each other. pretty darn good episode) Insofar as your original sentiment tho i fully agree and would add that I'd also like to hear Wolfram talking with levin, it's like levin is approaching organization from biological-cellular automata while bach is from digital-automata
What would JB do?
try to convince us to align purposes with awesome snarky insights ;)