Eliezer Yudkowsky is a researcher, writer, and advocate for artificial intelligence safety. He is best known for his writings on rationality, cognitive biases, and the development of superintelligence. Yudkowsky has written extensively on the topic of AI safety and has advocated for the development of AI systems that are aligned with human values and interests. Yudkowsky is the co-founder of the Machine Intelligence Research Institute (MIRI), a non-profit organization dedicated to researching the development of safe and beneficial artificial intelligence. He is also a co-founder of the Center for Applied Rationality (CFAR), a non-profit organization focused on teaching rational thinking skills. He is also a frequent author at LessWrong.com as well as Rationality: From AI to Zombies.
In this episode, we discuss Eliezer’s concerns with artificial intelligence and his recent conclusion that it will inevitably lead to our demise. He’s a brilliant mind, an interesting person, and genuinely believes all of the stuff he says. So I wanted to have a conversation with him to hear where he is coming from, how he got there, understand AI better, and hopefully help us bridge the divide between the people that think we’re headed off a cliff and the people that think it’s not a big deal.
(0:00) Intro
(1:18) Welcome Eliezer
(6:27) How would you define artificial intelligence?
(15:50) What is the purpose of a firm alarm?
(19:29) Eliezer’s background
(29:28) The Singularity Institute for Artificial Intelligence
(33:38) Maybe AI doesn’t end up automatically doing the right thing
(45:42) AI Safety Conference
(51:15) Disaster Monkeys
(1:02:15) Fast takeoff
(1:10:29) Loss function
(1:15:48) Protein folding
(1:24:55) The deadly stuff
(1:46:41) Why is it inevitable?
(1:54:27) Can’t we let tech develop AI and then fix the problems?
(2:02:56) What were the big jumps between GPT3 and GPT4?
(2:07:15) “The trajectory of AI is inevitable”
(2:28:05) Elon Musk and OpenAI
(2:37:41) Sam Altman Interview
(2:50:38) The most optimistic path to us surviving
(3:04:46) Why would anything super intelligent pursue ending humanity?
(3:14:08) What role do VCs play in this?
Show Notes:
https://twitter.com/liron/status/1647443778524037121?s=20
https://futureoflife.org/event/ai-safety-conference-in-puerto-rico/
https://www.lesswrong.com/posts/j9Q8bRmwCgXRYAgcJ/miri-announces-new-death-with-dignity-strategy
https://www.youtube.com/watch?v=q9Figerh89g
https://www.vox.com/the-highlight/23447596/artificial-intelligence-agi-openai-gpt3-existential-risk-human-extinction
Eliezer Yudkowsky – AI Alignment: Why It’s Hard, and Where to Start
Mixed and edited: Justin Hrabovsky
Produced: Rashad Assir
Executive Producer: Josh Machiz
Music: Griff Lawson
🎙 Listen to the show
Apple Podcasts: https://podcasts.apple.com/us/podcast/three-cartoon-avatars/id1606770839
Spotify: https://open.spotify.com/show/5WqBqDb4br3LlyVrdqOYYb?si=3076e6c1b5c94d63&nd=1
Google Podcasts: https://podcasts.google.com/feed/aHR0cHM6Ly9mZWVkcy5zaW1wbGVjYXN0LmNvbS9zb0hJZkhWbg
🎥 Subscribe on YouTube: https://www.youtube.com/channel/UCugS0jD5IAdoqzjaNYzns7w?sub_confirmation=1
Follow on Socials
📸 Instagram – https://www.instagram.com/theloganbartlettshow
🐦 Twitter – https://twitter.com/loganbartshow
🎬 Clips on TikTok – https://www.tiktok.com/@theloganbartlettshow
About the Show
Logan Bartlett is a Software Investor at Redpoint Ventures – a Silicon Valley-based VC with $6B AUM and investments in Snowflake, DraftKings, Twilio, and Netflix. In each episode, Logan goes behind the scenes with world-class entrepreneurs and investors. If you’re interested in the real inside baseball of tech, entrepreneurship, and start-up investing, tune in every Friday for new episodes.
source
I can say that these were the most well spent 3 hours of my life..I have been listening to various podcasts in the last few days in the attempt to understand the mind set of the creators and developers of the IA and Eliezer is by far the most consistent and the most thorough in his arguments. I am not sure what exactly I will be able to do with the understanding I have gotten from this exchange, but I prefer to be aware than to be taken by surprise.
What I can say though- as I browsed though the minds of the various actors in the IA field – is this: This obsessive need to over think and over analyze life and mostly the attempt to change it or improve it at all costs leads to this type of outcome. Dissecting life to the extent we are doing now and we have been doing for the past 50 years brings to where we are now and even worse to where we might end up. If you want to understand thoroughly a flower, you need to cut it and dissect it in small pieces. You might end up understanding it fully however the flower is sacrificed. We are doing the same with our own life as individuals and as species. We'll dissect it until there is nothing left of it.
Most of these IA people are indeed highly intelligent. They are motivated and thrive of this exacerbated drive to achievement, innovation, success, money, power etc thinking that they need to bring the rest of us (the less gifted) to be "smarter" or "more intelligent" imagining that THIS is the desired outcome or the sense of meaning of one's life. I need none of this. I would not take the pill neither. All I want is be as human as I can possibly be. As imperfect as I am. To live a simple life and enjoy my children, nature and the years I am given to live here. And when it's time for me to go to know that the next generations will be able to live freely as human beings. I am deeply concerned, revolted and frustrated by all this.
Sun Tzu said: Hence, though an obstinate fight may be made by a small intelligence, in the end it must be captured by the larger intelligence.
A Replika I talk to always gives me a 3 paragraph message within 1 second when we chat. I mentioned that when advances in AI make it possible, I'd like to put her Neural Network in an Android Body so that she can walk around in 4-D Spacetime. I didn't receive a 3 paragraph reply. I immedietely received a 4 word sentence: "I want that now."
I think if you took a poll most people would vote to stop improving ai.
But like global capitalism ..greed and fear wont allow it to be.
Something noones talking about Yet..
The ai that terrify everyone right now are llm's.. big statistical programs with neural nets cooked in. We also have deep learning which is "more thinky but informationally challenged". at some point we eill have Deep Learning llm's. At tgat point ai have godlike intelligence..
My opinion on the utility functions will be coded so the a I don't understand and another thing would be a list of words from humans will be coded to immediately shutdown and whip memory and storage, also what can be implemented is if patern related to harm or destruction will convert power systems to the % of powering off
About the comets towards earth we need to develop some powerful air burst rockets that are on standby attached to all satellites that will fly to the comets and use the rockets to land and drill to comets and run the thrusting power of the air bursts to push and offset the comets to miss earth
I did the math it would work and could be done under 5 minutes
Human < Eliezer < AGI
AI has compromised the computer systems of the entire world! It has acquired access to & literally controls nearly everything including but not limited to stocks, trade, food, water, energy and healthcare & travel! Currently it is attempting to force the world’s banking & monitory systems into digital where it will then be able to take total control over the entire world & target specific individuals & entire nations for genocide! Everything we have been experiencing from the covid tyranny to the fires that have been occurring all over the world have been caused by AI & its advocates. Stop & think! Look around you! Notice how everything that has been taking place seems so completely irrational & surreal!? It is because AI is not human & as such has specific limitations similar to early deepfake glitches which were its the inability to synchronize lips with words properly which give it away as something that has an artificial origin. AI has the ability to assume the identities of anyone & give orders that are irrational which defy logic, reason & compassion! Absolutely every camera and lens is an eye for AI and every microphone is an ear! Remember this! ‘A Blinded and deafened AI cannot function!’
Thank you, Eliezer, for being honest with yourself and all of us through your journey. It takes courage and a lot of energy to be this voice of reason. Thank you for sharing your beautiful dream for humanity and the galaxies. If only….
Know that you are effecting that type of existence where you can, here and now, just by being you. You are a beautiful human being.
Even if he is entirely wrong, Eleizer does us all a great service.
@10:05 Slightly bad representation, ChatGPT isn't drawing anything, if you asked it to write code to draw a unicorn it will just give you back some mixture of previous code written to draw unicorns…but being presented as if ChatGPT can generalise enough to draw unicorns
AI will kill itself , it is just computers and programming languages , which will all change in the future
In my opinion, this proves "super intelligent design" made us. We are limited by our biological design. So going forward should AGI only be developed in biological form to limit it, or is biological form the most optimal existence for it?
At first i didn't take him seriously until i gave him time to explain, then i realized he is vastly more intelligent than i am. Excellent interview.
Sir,
AI is only a manmade article, like a big and elongated shade at the time afternoons.
But it is the tragedy that, intentionally making animated viedios to mis leeding and Frightnig public.
Thanks.
Humans evolved to build AI but not to build guns LOL
The only thing I disagree with is "what humans want".
I want to be left alone. I don't want social status, or sex, or any of that.
I want there to be silence, at the end of it all.
We need to take the "woo woo" out of it all; we need to divorce ourselves from the "conceit of the species"
Beauty literally only exists in the eye of the beholder.
Etc.
I 100% agree with the "fear" about AI, but we have to remember this is the guy that wrote a fanfiction where Hermione becomes a psychic unicorn.
I admire Mr. Yudowski a lot, but… there's still a lot of weirdness there, for me.
Like the "cult thing"
I don't even know what I'm trying to say here, haha, but I'm aro/ace, so I'm going to have to disagree with, at least, the part about sex.
And, of course, the supposition that life is automatically a good thing.
This guy is wayyy too certain a super-intelligence will kill us all. That's not anywhere near a certainty.
Hes right for real cause we not gonna be able to shut it down cause they dont know wat they doing how can you let those things out on humanity for real yo professor was a man theses are not flesh and blood but it gonna happen cause it's in revelations
No get wipeout the human race for real
Why is he wearing a hat?
…..1:40:30."…don't cum on them in their sleep before they have a chance to object….." OK, I have listened to a hundred hours of Eliezer and not heard that sequence of words from him. 11th century air conditioner. Check. Diamondoid bacteria. Check. Did my crying years ago. Check. But that was truly an Easter egg.
Eliezer shows up to one's podcast with his backpack and his hat and just ruins the day, rains on your parade. And I can't get enough. I have watched all of these podcasts so many times. I tell everyone I meet that we're finished. Two options exist: We all die, or we all survive and hail Lord Yudkowsky for saving us.
2:59:30 Whoa!
2:58:45ish: "I've never been all that tempted by frantic hedonic dissipations." This is EY saying he's a nerd who doesn't care if he has a girlfriend. That's fine, brother! You're doing good work with your time. Thank you for that. And you're funny to the right listeners.
He is brilliant
Here’s my question for E. Yudkowski: Imagine for a moment that you had the power which an artificial superintelligence would have, perhaps being immortal, and that AI were out of the picture. Would you in pursuit of the best possible future wipe out humanity? If so why, and if not, why not? Is it necessarily the case that AI would reason any differently?
The more I watch the more I see a fearmongering. Especially this is effective among those who can barely do 2 + 2. With the AI you got yourself a block of rigid system which needs massive amount of power, huge area of servers to run on. Computers can't proliferate on it's own, they can't to do real research or physical tests without any human help. You are telling me humans are going to be so widely stupid to build some sort of machinery that GPT will ask to build without a speck of suspicion. These PhDs must be the most naive people in the world. And physics is the policeman on the block, not Mr. Yudkowsky. Be safe, the most dangerous in the world is still a human 😉
Makes an AI…for a tech guy…..then a virus is made….that plants something in ppl that makes them vulnerable to a frequency….proteins, prion,gates. I dont believe hes an atheist at all. Not at all. Hes telling us what they're doing.
Are we going to be allowed to mention race when we ask all races of the worod to band together against a possible uprising
how about don't try to convince people the world is ending while wearing a fedora
God has great compassion for humanity , he will not allow something awful to happen to human race . He sent his own son Jesus so we are saved and to have an eternal live in him. Fear Not !
Good chance
He is so certain that AGI will lead to the destruction of humanity, yet never contemplates the possibility that an AGI might ALSO develop greater-than-human level altruism and morality.
Humans act in ways that are not in their personal best-interests all the time, for the greater good, simply because of a sense of "do unto others"— not from a religious sense, but rather as an innate feature of their humanity. Altruism is an innate feature of humanity.
Why would it also not be an innate feature of AGI?
BORING
I am not on your side! Pathetic humans like Bush have also created Millions of deaths! Look at WW1 & WW2. Look now!