HN Theater @HNTheaterMonth

The best talks and videos of Hacker News.

Hacker News Comments on
Joscha Bach: Artificial Consciousness and the Nature of Reality | Lex Fridman Podcast #101

Lex Fridman · Youtube · 199 HN points · 11 HN comments
HN Theater has aggregated all Hacker News stories and comments that mention Lex Fridman's video "Joscha Bach: Artificial Consciousness and the Nature of Reality | Lex Fridman Podcast #101".
Youtube Summary
Joscha Bach is the VP of Research at the AI Foundation, previously doing research at MIT and Harvard. Joscha work explores the workings of the human mind, intelligence, consciousness, life on Earth, and the possibly-simulated fabric of our universe.

Support this podcast by signing up with these sponsors:
- ExpressVPN at https://www.expressvpn.com/lexpod
- Cash App - use code "LexPodcast" and download:
- Cash App (App Store): https://apple.co/2sPrUHe
- Cash App (Google Play): https://bit.ly/2MlvP5w

Incredible comment by ernst.gemeint showing some of the topics mentioned:
https://www.youtube.com/watch?v=P-2P3MSZrBM&lc=UgzzoW0z_tyeLyatEJB4AaABAg

EPISODE LINKS:
Joscha's Twitter: https://twitter.com/Plinz
Joscha's Website: http://bach.ai/

PODCAST INFO:
Podcast website:
https://lexfridman.com/podcast
Apple Podcasts:
https://apple.co/2lwqZIr
Spotify:
https://spoti.fi/2nEwCF8
RSS:
https://lexfridman.com/feed/podcast/
Full episodes playlist:
https://www.youtube.com/playlist?list=PLrAXtmErZgOdP_8GztsuKi9nrraNbKKp4
Clips playlist:
https://www.youtube.com/playlist?list=PLrAXtmErZgOeciFP3CBCIEElOJeitOr41

OUTLINE:
0:00 - Introduction
3:14 - Reverse engineering Joscha Bach
10:38 - Nature of truth
18:47 - Original thinking
23:14 - Sentience vs intelligence
31:45 - Mind vs Reality
46:51 - Hard problem of consciousness
51:09 - Connection between the mind and the universe
56:29 - What is consciousness
1:02:32 - Language and concepts
1:09:02 - Meta-learning
1:16:35 - Spirit
1:18:10 - Our civilization may not exist for long
1:37:48 - Twitter and social media
1:44:52 - What systems of government might work well?
1:47:12 - The way out of self-destruction with AI
1:55:18 - AI simulating humans to understand its own nature
2:04:32 - Reinforcement learning
2:09:12 - Commonsense reasoning
2:15:47 - Would AGI need to have a body?
2:22:34 - Neuralink
2:27:01 - Reasoning at the scale of neurons and societies
2:37:16 - Role of emotion
2:48:03 - Happiness is a cookie that your brain bakes for itself

CONNECT:
- Subscribe to this YouTube channel
- Twitter: https://twitter.com/lexfridman
- LinkedIn: https://www.linkedin.com/in/lexfridman
- Facebook: https://www.facebook.com/LexFridmanPage
- Instagram: https://www.instagram.com/lexfridman
- Medium: https://medium.com/@lexfridman
- Support on Patreon: https://www.patreon.com/lexfridman
HN Theater Rankings

Hacker News Stories and Comments

All the comments and stories posted to Hacker News that reference this video.
Same for me. His first interview with him was also my favorite podcast episode of 2020.

Also, you can watch it in video form. I'd argue it's better that way, especially with the way he emotes during certain parts of the discussion.

Episode 1: https://www.youtube.com/watch?v=P-2P3MSZrBM

Episode 2: https://www.youtube.com/watch?v=rIpUf-Vy2JA

Joscha Bach: Artificial Consciousness and the Nature of Reality | Lex Fridman Podcast #101

https://www.youtube.com/watch?v=P-2P3MSZrBM

julvo
Joscha Bach was for sure one of my favourite discoveries this year. He seems like someone who has deeply thought about many disciplines including CS, philosophy and psychology
tasuki
Aside from the very interesting content, I really enjoyed how Joscha sounded kind of perpetually amused.
Totally agree. Joscha Bach laid out a few interesting ideas around this at the end of a podcast.

If you have a few minutes start here for a broader context (starts of wholly unrelated but with some great concepts):

https://youtu.be/P-2P3MSZrBM?t=10084

If not, he drills into it here:

https://youtu.be/P-2P3MSZrBM?t=10400

"everything is downstream from controlling attention" - Joscha Bach on the Lex AI podcast https://www.youtube.com/watch?v=P-2P3MSZrBM&t=3s
biddit
That’s an amazing observation. Can you point to roughly where in the video he says this? Thanks!
sova
2:03:00 begins a discussion on meditation and at 2:03:45 he says "everything is downstream from controlling attention" -- toward the end of the segment labeled "AI simulating humans to understand its own nature"
heed
It's about here (2:03:40): https://youtu.be/P-2P3MSZrBM?t=7420
Sep 01, 2020 · miketery on Mind Emulation Foundation
The best I've seen the topic of mind and computation bridged and well communicated has been by Joscha Bach. Recommend watching this video on Computational Meta-Psychology [1]. Also his 3 hour conversation with Lex Fridman is a whole other amazing rabbit hole [2].

[1] https://www.youtube.com/watch?v=WRdJCFEqFTU

[2] https://www.youtube.com/watch?v=P-2P3MSZrBM

Jul 23, 2020 · 2 points, 1 comments · submitted by novalis78
Funes-
Already on the front page: https://news.ycombinator.com/item?id=23923183
Jul 23, 2020 · 186 points, 76 comments · submitted by TheAceOfHearts
uxcolumbo
What a coincidence - I was just watching this month old interview a few hours before it was posted here. Maybe Youtube's algo is pushing it...

Anyway, the way Joscha explained various concepts was very interesting - the phrases he used kept me hooked. At the very least it's entertaining.

For example:

"Social media is the world's brain hooked on dopamine - creating a permanent seizure"

"Entropic Abyss"

"Closed Cooling Chain"

Interesting guy with an interesting upbringing (grew up in Eastern Germany - his site http://bach.ai).

It's a long interview - but here is the breakdown of topics being discussed (time stamps taken from the comments section)

0:00 - Introduction

3:14 - Reverse engineering Joscha Bach

10:38 - Nature of truth

18:47 - Original thinking

23:14 - Sentience vs intelligence

31:45 - Mind vs Reality

46:51 - Hard problem of consciousness

51:09 - Connection between the mind and the universe

56:29 - What is consciousness

1:02:32 - Language and concepts

1:09:02 - Meta-learning

1:16:35 - Spirit

1:18:10 - Our civilization may not exist for long

1:37:48 - Twitter and social media

1:44:52 - What systems of government might work well?

1:47:12 - The way out of self-destruction with AI

1:55:18 - AI simulating humans to understand its own nature

2:04:32 - Reinforcement learning

2:09:12 - Commonsense reasoning

2:15:47 - Would AGI need to have a body?

2:22:34 - Neuralink

2:27:01 - Reasoning at the scale of neurons and societies

2:37:16 - Role of emotion

2:48:03 - Happiness is a cookie that your brain bakes for itself

EDIT: clarity

agigao
Thank you!
08-15
I noticed the same thing and had the opposite reaction.

"Social media is the world's brain hooked on dopamine - creating a permanent seizure"

Wow, what an insightful statement! Except it isn't. It's not even a statement. It's just different metaphors piled on top of each other until they nothing at all anymore.

"Entropic Abyss"

Lex Fridman actually asked for clarification here, but none came. Joscha fixed the mispronunciation, that's all. The phrase is completely meaningless.

"Closed Cooling Chain"

And that one is a bad translation from German "geschlossene Kuehlkette", meaning that your convenience food is frozen once at the factory and never warms up until it's in your microwave oven. It translates to "uninterrupted", not "closed".

I'm pretty disappointed by the onslaught of empty phrases and the dearth of insight or arguments.

uxcolumbo
That social media statement - doesn't necessarily have to be super insightful.

The mental image triggered by that sentence is funny and there is some truth to it.

> "Entropic Abyss" ... The phrase is completely meaningless.

Well, it wasn't meaningless to the interviewer - he actually found it poetic.

And for me it triggered an interesting mental image.

> bad translation from German... It translates to "uninterrupted", not "closed".

I just looked up the words. 'Geschlossene' comes from schliessen, which means to close.

Uninterrupted in German seems to be 'ununterbrochen'.

So Closed Cooling Chain actually seems to be a good translation and it nicely rolls off the tongue - CCC.

> I'm pretty disappointed by the onslaught of empty phrases and the dearth of insight or arguments.

I realized some might not get much out of it - hence why I said that at the very least Joscha is entertaining to listen to (subjective - I know).

I got something out of it - what Joscha says makes me think - even if at the end I won't get enlightened, I look forward to check out his other work.

08-15
> The mental image triggered by that sentence is funny and there is some truth to it.

Sure, but funny isn't insightful.

> I just looked up the words.

Words from one language don't map perfectly to those of another; idioms map even less well. Are you seriously trying to lecture me about idioms in my native tongue, German? Which happens to be Joscha's native tongue, too, which is probably the reason why some of his phrases sound insightful to you and like bad translations to me. Another example:

"A spirit is an operating system a brain runs on."

Apparently he tried to translate German "Geist", which could be "spirit", "ghost", or "mind". The appropriate translation is "mind", then it makes sense. The mistranslation to "spirit" makes it sound profound, unless you take the time to think about it. But you can't, because Joscha talks so fast.

uxcolumbo
> Sure, but funny isn't insightful.

Is that an established and commonly understood fact?

To me anything that provides some understanding to a complicated matter or situation can be insightful - even funny things.

What isn't insightful to you might be insightful to others. Don't forget not everyone is at the same level of understanding.

> Closed Cooling Chain vs Uninterrupted Cooling Chain

When he says 'Closed' - it implies that it's uninterrupted. Have you ever heard of the term 'Closed Circuit'? Same thing.

I prefer his translation. It's simpler and sounds better (to me).

> Apparently he tried to translate German "Geist", which could be "spirit", "ghost", or "mind"

Interesting. Did he use 'Geist' originally in his German presentation?

If he meant mind - why didn't he use the German word 'Verstand' - which seems to be a more accurate translation for mind. But you tell me, as this is your native tongue.

Maybe he wants to add some colour to his talks?

If so, then I think spirit is totally fine to me. It doesn't make it sound anymore profound. I would say it makes it more interesting. And maybe the other reason why he used the word spirit and not mind, might be to cater to a certain audience, e.g. the viewers for the chaos computer club videos might appreciate sci-fi like Ghost in the Shell. You wouldn't say Mind in the Shell. Sounds dry and boring (to me). But spirit has a more positive connotation vs Ghost - another reason why he might have chosen that word in the quote you mentioned.

Joscha seems to have a knack for this - explaining things in a simple, interesting and entertaining way. Just look at all the comments on the various videos he appears in.

You might already have a deep understanding of the topics he talks about, but not everyone has.

08-15
> Have you ever heard of the term 'Closed Circuit'? Same thing.

Of course I've heard of it! Very much not the same thing. The cooling chain is open at both ends. It isn't closed like a circuit, it is still called "geschlossen", which is exactly why I'm trying to tell you it's a mistranslation and "uninterrupted" conveys the actual meaning much better.

> why didn't he use the German word 'Verstand'

To sound "spiritual", literally, so he can connect with religious nutters instead of just nerds. What do I know of his motivations?

jmiskovic
I was quite intrigued by semi-proposal of building a meta social network (it starts around 1:37:45). He proposes a service where anyone can start a social network using any features of existing ones (Facebook feed, Twitter conversation, Reddit threads...), with each network cultivating its own rules and reward function (criteria for what content is reshared or upvoted).

On one hand it's over-ambitious and doomed to fail, and yet it could lead to more coherent and constructive hive mind.

cosmojg
Isn't that kinda what's already unfolding with ActivityPub? You've got Friendica for Facebook, Mastodon and Pleroma for Twitter, Lemmy for Reddit, WriteFreely and Plume for Medium, Funkwhale for SoundCloud, Peertube for YouTube, Pixelfed for Instagram, and so much more[1]! Not only can anyone spin up their own instance of any one or more of these services with whatever rules and customs they want, but all of these different services can talk to each other since they're all built on the same underlying protocol. It's cool stuff, and it's already happening today!

[1] https://fediverse.party/en/miscellaneous

cblconfederate
It has been tried so many times -- evolving variations of social networks. The end result is facebook bought them all. It's what we call survival of the richest.
yetihehe
> with each network cultivating its own rules and reward function (criteria for what content is reshared or upvoted).

Yeah, that sounds like trying to evolve even more addicting social networks. Those networks which are more rewarding will have more users, regardless if they are good or bad in long term.

rnd_74367843
I tried to do that 10 years ago. Didn't gain traction. Too complex an idea for most to want http://babblingbrook.net/
alexshortt
I really love his emphasis on modelling as a skill. Its obviously great applied to technical skills like computer science, but he seems to apply it to everything else in the human experience so well.

"everything that we sense is part of the same object ... we learn it all into one model ... we call this model the universe" https://youtu.be/P-2P3MSZrBM?t=1796

callesgg
Everything is a model. The only thing that exists to the mind And actually in the world is models.
calebm
As above, so below
procgen
Tat tvam asi
arethuza
Reminds me of "the world is its own best model" and the subsumption architecture of Rodney Brooks:

https://en.wikipedia.org/wiki/Subsumption_architecture

motohagiography
I'm such a fan after the 33C3 Machine Dreams talk.

The podcast host, either he has a way of speaking that seems affectedly slow, or maybe he has a speech impediment or neurological issue? Ordinarily I would just set the speed of the video up, but Bach's normal speed of speech becomes too fast. A few minutes in, the host seems to adapt to his guests speed a bit, but is there some kind of trend among millennials to do some weird slo-mo lazy talking pace? Bach seems to sense it as well and interrupts him when he starts to wind up his mental drone.

therufa
His first language is Russian, but grew up in the US.

From first hand experience as a person who grew up in a household with two completely different languages, this seems to come from having trouble of finding the right language to speak in for a little time until I can finally settle in on a language which makes most sense.

This or something absolutely different.

motohagiography
It would be surprising if it were a language issue, given Bach speaks accented english much more rapidly than the host speaking mostly unaccented english. It seems received or an personality/attitude thing, or maybe it's a substance use issue. Anyway, Bach is interesting enough I'm putting up with it, but I rarely feel so irritated and resentful of a person, so if someone else is repelled by it, stick with the podcast as Bach makes it worth while.
logarhythmic
It doesn't really bother me, but for what it's worth, the host (Lex Fridman) has stated that it's simply the way he speaks as a result of his brain trying to choose what thread to follow and what to say
motohagiography
Lex's comment toward the end that lawyers are the substrate of reality won me over. That's hilarious. Was distracted by his mien, but funny justifies all.
skovorodkin
For recording it's much better to speak slowly. You can always speed up slow speech, but doing the opposite (slowing down fast speech) does not bring you a good result yet.
ilaksh
Bach is rapidly firing out intriguing statements.

Lex is just trying to come up with worthy follow-up questions that also sometimes align with his plan for the podcast. And his default is to speak carefully because he knows that the topics are complex and algorithms are transcribing them etc. So he is starting at a fairly slow normal just for clarity and the high level of Bach's comments requires him to think more.

I would bet that you would not be able to do better. Actually I bet that you would pause just as long and then often reply with something barely relevant to what was just said, such as "wow that was deep".

Seriously, you think that Bach hasn't spent years preparing that elegant information and philosophy? And Lex was trying to adapt on the fly. But you casually suggest that you could have a rapid-fire intelligent back-and-forth with Joshua Bach as he lays that stuff on you.

I would like to see it. Go ahead, record a conversation with Bach on a topic that is slightly different from the existing podcasts/talks.

keeganpoppen
why does OP's interviewing skill have to do with anything? i actually agree that stilted non-conversational speech like that in interviews and speeches sounds bad (talking in particular about the beginning). no need to say all that live if it's not contingent on the interviewee in any way. just watch the video: he is sitting there staring at a piece of paper reading, and then the first question the interviewer asks is essentially "what are some of the foundational inspirational ideas of your work?", which for anyone who isn't brilliant and hyperrational like the interviewee would be difficult to answer after being warmed up for an hour, much less before even seeing the eyes of the interlocutor. yes, interviews are not the same thing as normal conversation, but i don't think it's that crazy to suggest (like OP) that priming people in a more conversational way can be a nice way to smooth the edges around a conversation that will already be difficult for everyone to grasp (including Bach, who has to articulate all these things).

all of this is independent of how good of work lex does as an interviewer, which seems like it's pretty good given his ability to engage with the ideas presented (in this context idea empathy (ie understanding what Bach's saying) > interpersonal/conversational empathy).

motohagiography
Sounds like I should do a podcast, thank you, I will seriously consider it. I think Bach is one of the funniest and most entertaining people I can remember listening to, as he's one of the few examples I know of who could meet the definition of a stand-up philosopher.

My ideal first episode would have Bach and Daniel Dennett together talking about whether or not they have dogs, and what they might perceive that relationship to be like, given their respective views of consciousness and language, and what their dogs might think of them.

jagrsw
I've seen the interview in June, and was following Joscha's twitter feed since.

Although a lot of good stuff, some of his views are a bit perplexing:

* If his offspring would turn homosexual, he'd be sad about them being "evolutionary dead-end" (tho, he'd still support them in any way possible) - cannot find the tweet now, he probably deleted it

* He expresses a dim view of US' future (its institutions) esp. in the light of current events. It's sometimes visible in his tweets and speeches. But he's so vague in his predictions, that 10 years down the road I wouldn't be able to tell if he have been right or wrong no matter what the situation will be

tangent point: the company he currently works for - AI Foundation - had some deal/interaction with our dear Deepak Chopra - who is as anti-intellectual as a public persona can be - so my question is 'why' - tho this could be before Joscha's tenure there... but still it feels bad (https://www.businesswire.com/news/home/20191205005164/en/Dee...) (https://aifoundation.com/author/lars_butler/)

But he loves Stanislaw Lem, and according to the interview he was the first author that he felt was worth reading his books had profound impact on him when he was young, so that's nice, b/c Lem is a genius - who is also a bit underrated in the western world - as opposed to the former eastern block :)

longtom
In one tweet he suggested the rise of social justice was in part fabricated to demolish the political left in light of Occupy Wall Street,[0] and that much of U.S. politics is about hiding who is actually in control. Funnily, he was indirectly funded by Harvey Weinstein.[1] Overall I think, he is extremely courageous and has perfected the virtue of telling the truth, like telling entire established buildings of philosophy that they are full of it. In my opinion, we need more of guys like him who simply tell the truth instead of chasing the latest political applause light.

[0] http://archive.is/wip/kvjGp

[1] https://www.campusreform.org/?ID=14814

masterproc
Given my comment above, I would completely disagree in this point. I agreed however, it seemed similar to me before the funding stories were published.
longtom
Oops, I meant Jeffrey Epstein. I wrote "funny" because of the ties to that U.S. elite and because he actually dares to indirectly question them all the time (but not via the fabricated social justice angle, of course). Also, the entire Weinstein/Epstein drama is an instance of sexuality having become absurdly politicized to the degree that mainstream knowledge on it is outright false.
honzzz
What is perplexing about it? Genuine curiosity - where I live those views would probably be considered mundane mainstream.
jagrsw
I replied on this in another subthread, a minute ago or so.

But as for the 'mainstream', Joscha exactly doesn't want to be mainstream - see his TT stream: :)

"In a playground where most people can neither come up with ideas of their own.." and various other tweets and presos related to the harm of conforming to societal norms of thinking.

antepodius
If your views happen to align with the mainstream, that still isn't a reason to change them in and of itself. Being ruled and controlled by the mainstream is bad, because the mainstream makes mistakes; but that doesn't mean that the right path is just to try and be as much the opposite of the mainstream as possible.
jagrsw
There's no disagreement in here.
throwaway69123
Why do you find him expressing sadness about him not being able to pass on his genetics, perplexing? Its a primal evolutionary imperative. While possible the rate of child having is very low in same sex relationships in comparrison to the hetro relationships is higher.
Barrin92
something being a primal evolutionary imperative doesn't make it any more meaningful. In fact it's kind of a stupid thing to care about because reproduction in that sense is truly mindless and circular.

Raising a kid is a meaningful experience, but the kids haircolor doesn't matter much.

Aeolun
The fact that the kid will have another kid and bring you grandchildren is though.
Barrin92
You can be gay and raise kids these days. That is to say Joscha Bach can given that he hails from Germany and the US, in some very conservative countries I guess that's an issue, but anyway that's not his point.
Aeolun
You can certainly be gay and raise children, but they’re at best half you and your partners’ biological offspring.
whack
If I told you that you can never eat chocolate again for the rest of your life, and you expressed sadness for not being able to experience the pleasures of eating chocolate, would that be equally perplexing?

Just because our desire for chocolate is a primal evolutionary imperative, doesn't mean that it's wrong to develop an attachment to it. Why is it ok to express sadness over missing out on physical pleasures, but not psychological pleasures.

Barrin92
because that's not the same thing. You can have the pleasure and meaningful experience of raising a child (which I remarked in my post), there is no actual pleasure or meaning to the idea of reproduction as such.

The actual comparison would be for me to express sadness about the fact that I cannot pass on my desire for chocolate to the next generation, which to be honest I couldn't care less about.

jagrsw
Mostly because he the object which creates the ideas and interacting with us is not his genes/body/brain but on the most abstract level it's his neural system, and the imperative you're talking about is imposed by his genes on his neural system (via brain wiring + hormones).

Since Joscha says that being adult means taking control of your emotions (I guess, to a point), so they serve your goals, and you're not at their whim, I wonder why he succumbs so easily to this specific emotion. Since having offspring has nothing to do with goals of his neural system (say... discovering the nature, advancing knowledge, exchanging ideas).

Goals of your genes != goals of your neural system (brain), and for the purpose of this conversation you are for me the latter.

blafusio
Because he is a human being.

And from a pure short sentence, you can't read anything out of it.

dan00
I'm pretty sure that's not what Joscha meant by taking control of your emotions. His point was certainly more in the direction of not unnecessarily suffering or creating harm by following your emotions.

Following your emotions regarding kids, helping them to raise and become good people, feeling fulfilled doing it, I can't see anything wrong with that. You could even see kids as advancing the knowledge through them. And if you feel that way about your kids, I can also see why someone would be sad not having grand children.

I wouldn't even call this sadness to be mainstream, because I think quite a few people don't allow themselves to feel that way or even condemn themselves. So it's quite honest to be open about such a sadness and shows a good understanding of yourself.

Kednicma
But children aren't obligated to pass on their genes in order to appease their parents. Parents only get to pass on genes for one generation, and get no further control over the process. To worry about having children because they might not have grandchildren is to be unfairly controlling over the future of children.
dan00
Feeling sad about it doesn't mean you're going to force your children having children.

You can have feelings without acting on them, but just perceive them. When Joscha talks about controlling your emotions this certainly would go into this direction.

jagrsw
In one of his appearances (maybe on the Fridman's show) he gave an example of being overly attached to outcomes of elections wrt to success of your favorite political party (being frustrated and sad when the party/president is not in power). His idea was that such feeling are not conductive to your wellbeing, and you should maybe do something about them, (just guessing: rationalizing it, meditating it out or somesuch).

I don't think being sad about reproduction choices of your children is beneficial, therefore I'd also suggest dealing with it. In the end it's just influence of your genes saying 'go, do something, talk to them, we need to be replicated'

dan00
> I don't think being sad about reproduction choices of your children is beneficial, therefore I'd also suggest dealing with it.

Feeling sad about something, doesn't mean you're sad about it all the time, that's suffering. You just can't control every emotion in you and there's also no need for it, it's just part of being human. Trying to control every emotion is certainly more harmful than accepting them.

> In the end it's just influence of your genes saying 'go, do something, talk to them, we need to be replicated'

Do you really think that why Joscha works in AI and his desire to expand the knowledge has nothing to do with his genes?

jagrsw
I have a favor to ask for participants of this thread. I'm genuinely curious, and I don't care my about my karma points.

What would be the primary reason for the proponents of the POV 'being sad about your grandchildren being gay b/c cessation of own genetic lineage isn't the best of approaches' being downvoted (like my every entry, and most others)?

I believe it's an interesting conversation, and both side of it exchanged some interesting ideas debating in good faith, being reasonably calm and avoiding common fallacies, so I suspect some factors that are not obvious to me, like maybe the interlocutors are worrying that even discussing such topic might be a danger to their future rights of having offspring it all, or somesuch similar (just a guess, probably not even close to the real reason)?

I'm trying to figure out what I'm being told, since downvotes don't come with comments/reasoning.

normallove
It's a recognized flag for a specific and rather imperial memeplex that is spreading rapidly. It also seems like a motte argument for the bailey of antinatalism, which is another strong and particularly distasteful -- for those who are not antinationalist I mean -- flag for this set of ideas that seems to come together or at least have an extremely tight correlation.
jagrsw
Yup, I agree with 'advancing the knowledge' through your offspring esp. if you consider your neural net capable and well conditioned (but again, who doesn't?:) and genetically+behaviorally inheritable, but he was talking more about grandchildren there (i.e. lack of them), and the effects will be much more limited in the third generation
andyjohnson0
He did pass on his "genetics" - he had children.

If he feels sad about them not having children, which is not really something a parent has a say in.

hackinthebochs
The game of evolution isn't to pass on your genes one generation, its to have your genes persist indefinitely. If his son chooses not to have kids then it counts as an evolutionary "failure" of the parent. It's why we choose attractive partners who will help us have evolutionarily fit children.
andyjohnson0
> The game of evolution isn't to pass on your genes one generation, its to have your genes persist indefinitely.

I think that's a maladaptive way to think about this. Attaching value and emotion to the long term fate of your particular genes is odd, given that they are almost completely identical to everyone else's. Whats the point in feeling sadness about this?

And evolution isn't a game. It doesn't have a point, or an outcome. It's just what happens when you have replicators in a free energy medium. Genes don't win or lose, they just make copies of themselves. It's nothing to do with _us_. Its just biochemistry.

antepodius
>they're almost identical to everyone else's

Do you really believe this argument? It seems like a strange one to make. Our genes are almost entirely identical, so why worry about diversity, then?

Besides, people like you will, all else being equal, surely get outbred by people who do care about their progeny. So why worry? The tendency to hold your point of view will wipe itself out...

jagrsw
> people like you will, all else being equal, surely get outbred by people who do care about their progeny. So why worry? The tendency to hold your point of view will wipe itself out...

That could be true (though, atheism is on the rise, despite religious people having more children than non-believers on avg). But it's also orthogonal to the question at hand. Existence or non-existence of proponents of a given POV at a given point it time doesn't say much about the validity of the POV itself.

antepodius
The POV's perfectly valid, because it's a set of values, which can't really be invalid per se.

But if I had the chance to pick my own values, I wouldn't want to have ones that were antimemetic.

andyjohnson0
> so why worry about diversity, then

Of course I'm concerned about that. Lack of diversity and the discrimination that goes with it causes real harm to real people.

> Besides, people like you will, all else being equal, surely get outbred by people who do care about their progeny. So why worry? The tendency to hold your point of view will wipe itself out...

So. Why should I care about that? If these people want to hold that view then that's up to them. What am I supposed to do: try to compete with them?

Whats the issue here?

hackinthebochs
>I think that's a maladaptive way to think about this.

It's by definition very pro-adaptive. Many if not most of our emotions and behavior is merely in service to passing on our genes (at some higher-order analysis), which you seem to claim is pointless. If being concerned about your lineage is pointless, so is everything else you presumably have no objection to, e.g. feeling sadness from being rejected by a romantic interest.

>And evolution isn't a game.

I'm sure you've heard of a metaphor before.

masterproc
FYI, Bach accepted money by Epstein after conviction. To me, someone who does not help Science, MIT and Harvard in fact finding about money by a convicted sex offender, is maybe not the best to read the future.

Sources: “The researcher Minsky had flagged for Epstein, Joscha Bach, declined repeated requests from me to discuss his ties to Epstein. But a 2018 paper on his theory of consciousness acknowledges support from the Jeffrey Epstein Foundation, and Bach has been listed in media tallies of Epstein grantees.” Science, At https://www.sciencemag.org/news/2019/09/what-kind-researcher...

“From 2014 to 2016, Bach’s primary affiliation was with MIT’s Media Lab. Harvard never paid Bach or provided funds to support his research, and no funds donated to Harvard supported Bach’s work. We understand that in 2014 and 2015, Bach’s work at MIT received fi- nancial support from Epstein.” (...)

“Though Harvard never paid Bach, PED’s website listed Bach as a PED research scientist beginning in 2014. We understand that Bach left MIT in 2016, but PED continued to list him as a PED research scientist until 2019. During that time, Nowak continued to give Bach access to office space at PED, which he used intermittently, and Bach often met with Epstein at PED when Epstein visited. PED has no records of Bach’s executing a Visitor Participation Agreement. Bach did not respond to our voice mail or email asking to speak with him, and we were unable to determine whether Epstein provided financial support to Bach after Bach left MIT and while he was associated with PED.”

REPORT CONCERNING JEFFREY E. EPSTEIN’S CONNECTIONS TO HARVARD UNIVERSITY. At https://ogc.harvard.edu/files/ogc/files/report_concerning_je...

“Epstein’s $100,000 donation in May 2013 was intended to be used at Ito’s discretion. His donations in November 2013 and in July and September 2014, totaling $300,000 (40% of Epstein’s post-conviction donations), were made to support research by Joscha Bach, a former Media Lab research fellow from Germany whom Epstein introduced to Ito in 2013. The Media Lab hired Bach in large part because Epstein subsidized the cost.“

* Bach declined to be interviewed in connection with Goodwin Procter’s fact-finding.

REPORT CONCERNING JEFFREY EPSTEIN’S INTERACTIONS WITH THE MASSACHUSETTS INSTITUTE OF TECHNOLOGY. At http://factfindingjan2020.mit.edu/files/MIT-report.pdf?20011...

NotUsingLinux
Thank you for sharing! Great interview and perspectives
None
None
deeplearning0
do they discuss also AI ethics? Wondering about funding ethics for research ...
magnusmagnusson
AI 'ethics' is how you get China dominating AI research eventually.

All these OpenAI spooks are either doing it intentionally or accidentally and at stake like this, what does it matter when the end result is a convincing loss.

yboris
I think you misunderstand what AI ethics is.

Do you want facial recognition cameras on every street? That's what no ethics oversight gets you.

antepodius
Ethics oversight won't stop there, though. The managerial class will seek to keep expanding it more and more, with increasingly arcane and insulting restrictions.

And is the research level the right place to stop that? I don't think so. That's a policy question.

yboris
There's a lovely analogy I like: driving fast.

New technology is like putting more weight on the gas pedal. Until we learn to handle the current problems, accelerating isn't a guaranteed solution. We may be able to swerve out of the oncoming problem, but now we're going event faster.

We now have technology that can end human civilization (aka "existential risk"). What is the harm in going slower? Especially compared to the harm of going faster.

antepodius
It's a coordination problem. Sure, you could slow down, and then the chinese will just outcompete you/spawn a paperclip maximiser- or, in general, a less scrupulous competitor will.

You'd have to get all sides to agree on an actually effective treaty.

magnusmagnusson
My sides are absolutely gone, what a hilarious man.
ExcavateGrandMa
The day a machine is gonna try to have a conversation with me I'm gonna smash it the concrete way! keep dreaming about the perfect creature/slave/SEXBOT :D
mrmonkeyman
You are just a memory of an AI trying to understand itself.
dane-pgp
ELIZA: That is interesting. Please continue.
ExcavateGrandMa
ELIZA??? all been already said... humanity isn't complex just ashamed by its own moves :D

and yea we are speaking about consciousnesss OH MY HEAD...

That's too much for me, can't comment much... but keep doing I'm watching... and guess what it's fun! :D

maze-le
He is definitely one of the most interesting people to listen to when it comes to AI, Intelligence and Philosophy. I've seen him in Leipzig on CCC events in the past, here are some of his other talks:

* Computational Meta-Intelligence (32C3): https://www.youtube.com/watch?v=WRdJCFEqFTU

* Machine Dreams (33C3): https://www.youtube.com/watch?v=K5nJ5l6dl2s

* The Ghost in the Machine (35C3): https://www.youtube.com/watch?v=e3K5UxWRRuY&t=2516s

He also did a very interesting podcast episode with Fefe and Frank from alternativlos! (in German): https://alternativlos.org/42/

andybak
I can't absorb complex ideas very well from speech. Any good links for written materials that offer good bite-sized introductions to his ideas?
moobsen
I'd recommending digging through his website: http://bach.ai/

Depending on what interests you, maybe also look at the abstracts of his research papers?

http://bach.ai/publications/

andybak
Thanks. His tweeter feed is quite interesting also (which gives you some indication of my attention span at the moment)
pantalaimon
His talks don’t have that level of snark he exhibits on Twitter, I guess that comes with the platform.
pantalaimon
I'm happy to get my annual Joscha Bach fix seeing that CCC is probably cancelled this year.

His talks are always a firehose of fresh ideas and new combinations of known concepts.

All of his previous talks can be found at

https://media.ccc.de/search?p=Joscha

Jul 21, 2020 · 2 points, 0 comments · submitted by tosh
Jun 21, 2020 · 2 points, 2 comments · submitted by vletal
vletal
I found this interview very interesting. It is rather long but Joscha Bach had something to say to plenty of vastly different topics with some pretty unorthodox ideas. I was wandering whether there is someone here with an interesting impression or critique to share.
magnusmagnusson
Jim Keller interview in same podcast is great too. https://www.youtube.com/watch?v=Nb2tebYAaOA
Jun 16, 2020 · 2 points, 0 comments · submitted by jonbaer
You might be interested in watching this: https://www.youtube.com/watch?v=P-2P3MSZrBM

If watching this isn't worth anyone's time, I'm not sure what would be.

throwaway9482
Wow the comments are giving this video really high praise. I’ve added it to my to watch later list. What did you find interesting in this video? Even after reading the comments I still am not quite sure what they go over in the video
AnimalMuppet
Sorry, I don't go watch random Youtube videos with no other information than "it will be worth your time". Mind telling us what the video is about, and maybe even why it will be worth our time?
throwaway9482
This sounds pretty rude FYI, I actually ended up asking effectively the same thing, you can look at my comment to see an example of how to phrase it more politely
AnimalMuppet
Well... I consider that posting a bare link with no description to be rude. The poster doesn't take the time to tell us anything (or, in this case, anything useful), while expecting us to take the time to go find out what it is. But there's one poster and maybe thousands of readers. To waste thousands of peoples' time to save your own is... let's say it's not very polite.

So, coming from that perspective, I am inclined to respond, um, less politely than you did. But I suppose than dang would encourage me to be better than the person I'm replying to, rather than just like them...

throwaway9482
Lol you’re not being a better person when you add a passive aggressive paragraph at the end but it’s all good, we’re on the Internet after all

And OP went out of his or her way to post that link I don’t think any of us are entitled to a summary or whatever

cuspycode
I didn't find it rude at all. The comment that presented the link was pure click-bait. I prefer it when a brief explanation accompanies a link, especially when the link points to a 3 hour long video, as in this case. The YouTube teaser says "explores the workings of the human mind, intelligence, consciousness, life on Earth, and the possibly-simulated fabric of our universe". I certainly agree that the topic sounds interesting.
throwaway9482
If the topic sounds interesting it‘s not click bait, it’s a good link that OP went of their way to share, we’re not entitled to anything
cuspycode
I didn't know anything about the topic before I clicked the link. Nothing in the comment that introduced the link revealed anything about the topic. And I only clicked the link because I was intrigued by the discussion here about rudeness. I was not motivated to click by the suggestion that watching would be worth my time (without any further details). That I found the topic interesting-sounding came after I swallowed the bait and clicked. Such things happen sometimes. And I still don't know if it actually is interesting, I won't actually know until I have set aside 3 hours to watch the video.
He recently did an interview with Lex Friedman (https://www.youtube.com/watch?v=P-2P3MSZrBM) and it was pretty good. It's a bit long but worthwhile. Covers a bunch of topics about the philosophy of machine learning, models of thinking and consciousness, and some bits and pieces of future directions for AI.
HN Theater is an independent project and is not operated by Y Combinator or any of the video hosting platforms linked to on this site.
~ yaj@
;laksdfhjdhksalkfj more things
yahnd.com ~ Privacy Policy ~
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.