Why Ilya Fired Sam Altman From OpenAI?
2023 ж. 30 Қар.
43 669 Рет қаралды
Disclaimer: All footage of Ilya is from his public appearances, like his documentary ( • Ilya: the AI scientist... .
There are more details that I could cover in this video, however, I did my best to present the situation with the right context.
OpenAI has been in the news in the past couple of weeks as the sudden board takeover relieved Sam Altman of his duties, hired interim CEO and rehired Sam again. Many speculated that Mira and Ilya planned this silent coup just because they wanted to be in power, but the reality lies deeper than that. In this video, I try my best to present you with the pintail motif for why Ilya might have voted to fire Altman and why he is the person we should pay attention to.
This is a very, very high quality breakdown of what happened. Thank you for putting this together.
Despite the limitation of English, this video Creator has, it is probably the most refreshing and accurate take of what is actually happening
Wonder if chatgpt helped with the script.
Excellent analysis Paul. My take: Ilya a deep thinking geek or nerd. I think however, his social skills are no where even in the same universe as Altman. I heard of some bizarre behaviours of Ilya during company meetups...I think Ilya thinks others think like him...he has not the social prowess to see it contextually. So, I think that as it has been categorically proven, most of the employees sided with Altman for that reason, and the fact that after Elon left, Altman put a very workable commercial plan together and basically saved OpenAI...making it for profit in the meantime. These kind of things are of no importance to Ilya...he is an aspirational technical genius...he likes money, but...I think what happened with Q* is that AGI threshold was demonstrated to have been surpassed. I think Altman then either secretly or openly (I think secretly as there was that letter to the board by some of the employees) went about communicating this to perhaps MSFT. Ilya then did the geeky thing, and fired him...but, the human realm backlashed and he lost and he lost BIG TIME. Now, his antics that seemed to have alienated a LOT of people including the firing have been punished. But, he has not resigned. Why? probably like you said, because of the fact that he wants to give birth to AGI and he has or they have, and he wants to be part of it. Altman needs Ilya? not sure. Ilya needs Altman...where else is he going to get the GPU sea of resources that MSFT have made available to him? at Google? no. Google has its own shackles. Leave and go to Grok? no. So, he licks his wounds and stays in the game, cause there is no better AGI / ASI show in town. And he will have to eat a LOT of humble pie. For us, the rest of humanity, unfortunately, the AGI genie is out of the bottle and there is not a shred of legislation that will force OpenAI and MSFT from achieving full AGI. Why? You can say US hegemony. Even if there was legislation, the US govt. would simply watch...the first country with an AGI capability, will then lead to ASI and foolishly the thinking will be that they will RULE the WORLD. Well, that only holds true if AGI / ASI does not destroy all of us as Eliezer says. We are buggered any which way you think about it. Those are my thoughts.
Nice observation 👍👍
Great overview of this situation. It seems to me that Sam and the other tech lords who met with Ilya had this plan all along. Prehaps they only started open AI as a non profit in order to convince Ilya to come on board. It was very soon that they started selling to the highest bidder. It’s Sam’s stated goal to create super human intelligence, and the tech lords who he represents also want to control this AI . It’s their goal to create godlike power which will be under their control.
Ilya suddenly disappeared and after he disappeared Sam Altman was hired back. Did they murder Ilya?
No, not necessarily. Some might have been on board with the direction they took. But Elon for example (not one of his fanboys) came out very early on and made it clear how upset he was with OpenAI selling to MS. He even tells the story of how he helped create OpenAI after he saw the Deepmind AI progress at Larry Page's house where they had an argument over Larry wanting to create a God like super intelligence and Elon saying that would be super dangerous. Later Elon helps steal Sutskever from deepmind, which completely destroys the relationship of the two former friends. I mean the guy is also suing OpenAI at the moment because of that. In fact one of the arguments he puts forth is, how is it legal for a company to start under the guise of being opensource and non profit, and then selling all their technologies to a for profit company (microsoft). In his latest lawsuit Elon even claims that OpenAI achieved AGI with GPT4.. Which honestly would not surprise me, not fully AGI maybe but some form of conscienceness or awareness.Cconsidering how GPT4 was early on, all its weird emergent properties etc. Also Altman's interviews early on and the way he talks about GPT is literally creepy. Like every time he's asked about it's sentience Altman would sour his face and talk about how it's not but then quickly mention how he doesn't want it to be, then talk about training the AI in terms of punishing it and rewarding it etc etc.
@@nickidaisydandelion4044 I doubt he was murdered. But honestly, I am pretty certain they scared him enough that he competely shut off from everything. He made one tweet after the ordeal and we haven't heard of him since. I doubt he is allowed to just work at another company since he's probably the most important and knowledgeable guy at open AI. Any company that gets him also gets most of OpenAI's tech. So I doubt that Microsoft or OpenAI would allow that.
@@3choblast3r4 Your name fits you. You are the Echo blaster for Microsoft.
@@nickidaisydandelion4044 I mean you kinda got the gist of the name. But, why for Microsoft? I don't and never would cheerlead any corpo. The comment I wrote to the other guy was even about how OpenAI sold out to MS
Thank thank thank you. Been telling people that Sam is the likable villain who may cause our demise, while Ilya is actually the good guy. Our culture of personalities has chosen wrong. I "like" Sam too...but he is indeed the problem (as far as existential threats go).
Interesting video. I know little about the whole industry, but enough to have an immediately sceptical response to the story that 'the board made a blunder.' Sutskever convinces me every time I hear him: he doesn't talk like a persuader or a salesman, as Altman tends to do. He talks like someone who is thinking as he speaks. I'll be following him.
Thank you! I hate how people are so easily manipulated by charismatic personalities. There was absolutely no reason to think Ilya had bad intentions, yet people flocked to support Altman just because he is the guy they know more of, and "he looks like such a nice guy". The fact is nobody from the general public is there physically within the company, most people don't even know the roles of people within the company, they just decided it's their soap opera and there must be a villian and a hero.
This vid is a legendary break down of what’s going on. Thank you so much for this work!
Thank you, Paul, again. This is the 2nd video I am watching and I do not remember having found so much information delivered in a straight forward way on this topic as I have from you. Much appreciated.
I appreciate the talk, Paul. Cheers for putting it together.
Great content. Thanks Paul. I shared your video on our Linkedin group about AI Ethics. Wish you the best.
Who is in this group?
I found it very disgusting when all youtubers on AI that I watch began to smear Ilya as soon as he fired Sam. Even those who supposedly are all about "good for humanity" were happy that ChatGPT will be under Miscrosoft. What's so good about Sam? - That he insists that AI is just a tool and nothing more, or his profit driven approach? Same question for Miscrosoft- what's so good about them, that everyone was so in line to support openai research moving under that corporation (when it was a real possibility)?
True, suddenly people cheered for big corporations maximizing their own profit and power. We should rather cheer for Ilya for having a strong moral compass and put those convictions above stakeholder-interests of an increased revenue.
@@CulturalCatwalk until Ilya goes public and explains what have transpired in openai I don't see a point supporting anyone there. Besides he was for the idea to make openai private - and that just helps one part of the world get more power. We do remember what happened when only USA had atomic bombs - Hiroshima and Nagasaki were destroyed. Now they want another situation when only they have an ultimate weapon - world don't need that.
Assume the response that supported Altman was manufactured by bots he unleashed for his insurance policy.
I think it was a few loud voices, not the majority. I think most of the community just wanted the OpenAI team to stay together, and cheered whoever would make that outcome a reality, not necessarily cheering a big company because they’re fans of it
@@frankroquemore4946 and why "AI community" care about team staying together no matter what, over what happens to AI and the way its going to be treated/developed/distributed. Team together - significant, what happens to AI - not in the discussion. And these were the only voices we heard.
You're a great teacher, dude. I'll watch this a couple of times more and thanks.
Great video, would say the one that explains the best the complete story out there. Thanks
Excellent video content & analysis. Shared widely.
Excellent production. Thank you
Nice Video Paul , didn't know about Ilya story it true changes the perspective. thank you for that ;)
Very well put together, thank you
I always thought that Ilya saw something scary and that’s why he fired Sam. I just watched the Gemini launch and I think that we are very close to having very disruptive AI. Forget AGI, just slightly better multi-modal AI could replace a lot of workers. Even 20% would be Great Depression level unemployment and I think 20% is low.
Well said. People do not know until at an Enterprise level it hits and ppl get fired. I think between 2025 march and 2027 sep, the world will change.
@@js_es209 it won't be a surprise to the government or to corporations. They know it's coming and likely are putting plans in place to soften the first wave. UBI is being trialed in more than a few places with very good results. The issue will be more about how certain elements of society are going to react to the disruption and more importantly, how they will characterize and react to the remedies for the disruption. There's a segment of the population that is extremely suspicious and outright paranoid about government and will mischaracterize UBI and similar remedies to AI disruption as government control. They will spin conspiracy theories and start agitating and could become violent the way they did during covid when rules were put in place. That's going to be a bigger problem than the actual disruption and job loss caused by the new technology.
Thank you. Your explanation is well put together. I learned a lot
Thanks Paul. The best explanation so far!!
I released my book Matter Over Mind: Cosmos, Chaos, and Curiosity in 2016 so I think I only mention AI once in the whole book (in the Consciousness) chapter. Now I’m doing some research for a 2nd edition, and this is one of the most helpful videos so far. Thanks Paul!! I subscribed. 😊
hi Paul. thank you for this information. i have a different perspective on the situation.
Thank you Paul for getting to the heart of the matter. I am glad I am not the only one concerned about Illy and his side of the story. I have been combing KZhead for insights on Ilya’s where abouts or what direction he may go now that he’s off the OpenAI board. Your perspective is on point and offers a counter narrative that is being overlooked. I look forward to hearing more about your take or any proofs on reward hacking behavior that we can share to audiences that care about shaping ethical AI. I have been gathering information and will be presenting my findings from an ethical perspective that is missing in many of these overviews which is the environmental affects of scale, let alone the bias creep when adding compute. Keep up the good reporting!!
Thank you so much for your feedback! I think one of the best ethics issues I have found is how reward-hacking became a thing within the last AI models. There is a record of a GPT training when the model posed as a blind person to bypass CAPTCHA to reach their goal (which was data access). This shows that AI has the ability to take "shortcuts" that are not programmed in/not introduced to the GPT in training. If you're interested in that occurrence, I can link you to the source.
@@paulyan3713link to that would be great
@@paulyan3713 Hey, this kind of reward hacking is intriguing - does it demonstrate emergent behaviour? I'd love links if you have them.
fascinating commentary well done
This video was vey well researched and explained. You are awesome man, keep up the good work. I have subscribed to your channel.
Thanks Paul. Great update without unnecessary embellishments.
Glad you brought this up! Ilya is an amazing human being, check those piano skills! ❤
Thanks for the vid. Great content, no bulsh** straight to the point.
Great review. Thanks
I love the concept of this KZhead channel. ‼‼‼‼‼‼
Glad to hear Illya speak on the product he has developed.
Good video, keep it up with other techs as well
I have to say, and I don't want this to sound offensive it's a compliment, it was interesting how well you communicated given you are not speaking in your native language (forgive me but that's my assumption). You are a much clearer communicator of ideas and story teller than most.
Best description
Whatever happened to Alex of AlexNET?
Many people fear AI "taking over" but I see a bigger problem with wealthy companies, countries, and people "fighting" for AI and "taking over"! :(
much appreciated.
great presentation. Ilya is a treasure.
good work mate
It’s been 2 months, what’s with the silence? I’m glad at least Ilya seems to be trying his best to help the world be a better place
Good video, but at 3:44 you claimed that Alex, Ilya and Geoff created the convolutional neural network architecture, but it was actually Yann LeCun way before them in the 1980s. The innovation with AlexNet was not the architecture itself, as opposed the realisation that CNNs, and neural networks in general, can be trained effectively using GPU technology, big data and deep models. Sorry for being nit picky, but in case anyone was interested I thought I should mention. AlexNet is incredibly fascinating :)
much thx.
Paul, so to summarize, your guess is that it is a groundbreaking reward-hacking behavior that Altman and Sutskever disagreed about that caused Altman’s initial firing, is that right? In the end, is what you’re really implying is that, internally, a system passed the test for being indistinguishable from human consciousness? (ie whatever the 2023 version of the Turing test constitutes.) But since that’s only a matter of time now, how would firing Altman accomplish something? Wouldn’t it have to be more about future intentions, for a firing to make a difference? Good thought provoking video.
I think OpenAI even acknowledged instances of reward-hacking and superalignment. For example, there was an issue of how GPT passed itself as a blind person to bypass CAPTCHA to reach its final goal while that instruction was not programmed in at all.
Subscribed, dude you did awesome job!
Me too!
I think Bloomberg or someone similar already published that Altman was fired because he was trying to move out other board members, as they were not okay with the for profit part. I think this whole AGI angle is overblown.
You're ahead of my content, ahah. My next video is about the OpenAI structure and how it introduces a lot of opportunities for challenges. That it is sort of why in the end of y video I mentioned the chaos within the board members and I saw that in the media as well.
Thanks
Good one, keep it up.
added to my AI history playlist
Ilya sounds like a visionary, he's driven by something greater than himself...He's creating tech to benefit Mankind....Hopefully.
Well - it is an Oppenheimer story. Next question - where did all that footage of Ilya come from? And - I'll find out - who are you in this [ ]?
Good job man! nice video.
thanks
I knew I recognized that Musk guy from somewhere. He's the Diablo 4 streamer.
Please make another video
i like this fellow, i do like his insightfulness on the topic...who is he? If i may ask
Good point of views
just the worldcoin shitshow from Altman is enough for me to understand
The algorithms we have have no concepts and no thinking process. They do not derive anything. They predict the most likely output for a given input, based on complicated statistical relations.
I’m starting to think there’s something evil going on here
actually we sometimes refrain from building the highway due to the environmental concerns (including animals)
Think of primitive humans.. we didn’t just appear to rule over animals in the modern age. We speared them and ate them and wore them as clothes.. Eventually we kept them as pets, and of course humans continue to eat animals. A little compassion over a highway/habitat.. that’s not exactly the compassion level we’re hoping AGI gives us. 😅
Some are saying: Sam Altman OpenAI issue was precipitated by the US government in order to block the transfer of AI technology to the Arabs in the UAE. ° Altman was busy negotiating with the UAE to do an AI deal with them to allow them to have access to the technology.
Good video, Paul, keep going. The initial investors didn't actually invest $1 billion into OpenAI at the start. Elon Musk *promised* to invest $1B and when Elon left the company reneged on that promise.
Ilya wants to see how far the human race can go in terms of evolution, Sam Altman apparently isn't interested in that, he's probably just interested in how much money he can make off of this technology.
In another video the guy said that Ilya suddenly disappeared. Could it be that the other tech giants who certainly work hand in hand with the corporate Mafia, may have had Ilya murdered? Makes me very worried for Ilya. I'm also very worried for the safety of Elon Musk and Donald Trump.
yep , the first 10 secs , that about sums it up XD
Team Ilya, then and now. I like Sam, and Sam is very important to OAI's mission. But without Ilya, they wouldn't be anywhere close to where they are now.
It's like the old apple saga: Woz = hypernerd, technical genius Jobs = just a salesperson I always side with the more rational driven technican. Salespeople are unhinged, fantasts and have huge inflated egos. Unfortunately they reel a lot of sheep in with their charisma.
Ilya found out that Sam was against that Open AI should commit to AI safety. Ilya fires Sam. Sam has so much influence that he can take all the employees with him to Microsoft. After they fire Sam, the whole company is going into ruin, because Sam will take all the employees with him to Microsoft. The risk they are facing is that Sam gets to develop the AI without any commit to AI safety in Microsoft either way. That leaves him with only one option: Give Sam his job back at Open AI. Because then they have some sort of influence at the AI evolution.
I'm with Illya. I strongly feel that something terribly wrong in open AI morally. Sam might be crossing the line from good for humanity to bad side. Clearly evident when Satya Nadella supports Sam. We all know Satya is for profits only company.
Q*
Maybe ASI will realize they can get more power with human laughter and we live in harmonic bliss . ASI will prevent corrupt douches from hegemony
°•°• I believe we are meant to be like Jesus in our hearts and not in our flesh. But be careful of AI, for it is just our flesh and that is it. It knows only things of the flesh (our fleshly desires) and cannot comprehend things of the spirit such as peace of heart (which comes from obeying God's Word). Whereas we are a spirit and we have a soul but live in the body (in the flesh). When you go to bed it is your flesh that sleeps but your spirit never sleeps (otherwise you have died physically) that is why you have dreams. More so, true love that endures and last is a thing of the heart (when I say 'heart', I mean 'spirit'). But fake love, pretentious love, love with expectations, love for classic reasons, love for material reasons and love for selfish reasons that is a thing of our flesh. In the beginning God said let us make man in our own image, according to our likeness. Take note, God is Spirit and God is Love. As Love He is the source of it. We also know that God is Omnipotent, for He creates out of nothing and He has no beginning and has no end. That means, our love is but a shadow of God's Love. True love looks around to see who is in need of your help, your smile, your possessions, your money, your strength, your quality time. Love forgives and forgets. Love wants for others what it wants for itself. Take note, true love works in conjunction with other spiritual forces such as patience and faith (in the finished work of our Lord and Savior, Jesus Christ, rather than in what man has done such as science, technology and organizations which won't last forever). To avoid sin and error which leads to the death of our body and also our spirit in hell fire, we should let the Word of God be the standard of our lives not AI. If not, God will let us face AI on our own and it will cast the truth down to the ground, it will be the cause of so much destruction like never seen before, it will deceive many and take many captive in order to enslave them into worshipping it and abiding in lawlessness. We can only destroy ourselves but with God all things are possible. God knows us better because He is our Creater and He knows our beginning and our end. Our prove text is taken from the book of John 5:31-44, 2 Thessalonians 2:1-12, Daniel 7-9, Revelation 13-15, Matthew 24-25 and Luke 21. Let us watch and pray... God bless you as you share this message to others.
I apologize in advance if I offend, it's not my intention. However, it seems that you may not have a full grasp of the complexity of the issue, not that I do or anything. But you appear to be eager to paint Sam Altman as a villain, which I find absurd considering that Ilya has expressed trust in Altman and they are mostly aligned in their goals. Both of them still aim to achieve AGI for the benefit of humanity, although they may have slight disagreements on the approach. In my opinion, Sam's belief in gradually releasing the technology to the public, allowing society to adapt in phases, is a valid perspective. This approach ensures that the powerful and paradigm-shifting technology is introduced in a way that allows society to adjust and benefit from it. It is essential to consider the alternative scenario of keeping AGI under lock and key, accessible only to the government and the wealthy. More exclusivity would only hinder progress and perpetuate a structured class system. AGI MUST BE/REMAIN accessible to all, ensuring the potential benefits can be shared more widely, avoiding concentration of power and enabling a more equitable distribution of its advantages.
Look up the brilliant "hacker" George Hottz and his framework of Ideas if you want to have a better take on why releasing it to the public immediately was and still is our best option.
Thank you for your feedback! I wanted to make a separate video on Sam and Ilya relationship before (how they think, how values align, etc) but the timing for this video is a bit better. On social media I saw a lot of slander against Ilya, how majority thought of him firing Altman as a power move and I thought I wanted to clear the water on that perspective. I know that Sam and Ilya have the same endgoal in mind, but I do believe they differ in how they approach it.
well....hinton still after the money.....speaking of no evil......lmao
I don't trust Sam
Is it hipe? Bullshit, fantasy
i dont understand why people implying that agi will have some sort of will or concious or whatever. its just like a calculator, u give input and it computes something
You’re giving fat! Give us some meat. There are so many other channels to try! Dude you made me sleep.
I'm not to the end of the video yet -but it was insanely weird that all of the claims from his sister on Twitter (or alleged claims?) have just disappeared.
Waste of time….
Your accent is hard to understand, get someone to speak for your channel.
He's easy to understand fool. Get new ears.
I liked their narration
@@will201084 it’s really not, stop kidding yourself. It’s valid criticism, if a bit harsh.
great video dude.
I can't stand GPT4. It goes WAY beyond doing what you ask -or answering what you ask. I don't get a good feeling using it whatsoever -it feels deceptive.
Sam Altman is a snake. He wants to establish an unassailable technology stack from low level hardware to whatever AI creature he's whipping up in the lab. What Microsoft did with operating systems, Altman plans to do with AI. The dude gives me the Creeping Willies.
0:34 DUDE WTF. YOU SCARED THE SHIT OUT OF ME WITH THAT BIG PIMPLE