What this covers

This is a 31-minute conversation, recorded at an Intelligence Squared event in September 2023, between Mustafa Suleyman and Zanny Minton Beddoes of The Economist. It centres on his book The Coming Wave. Suleyman co-founded DeepMind and Inflection AI, and he opens with some of that history: Geoffrey Hinton was DeepMind's first paid advisor, and several of OpenAI's co-founders passed through the company. He then sets out how deep learning taught models to interpret raw, messy data, and why compute for frontier models has grown roughly tenfold every year for a decade.

The middle of the discussion covers technical forecasts and risk. Suleyman expects AI to move within five years from producing single outputs to planning across multiple time horizons, such as researching, designing, negotiating and marketing a product. On containment, he notes that large models are trained on internet data that includes weapons-manufacturing instructions. Major labs can screen this out, but he argues open source models trail the closed ones only slightly and will catch up, so safeguards at the big labs will not hold everywhere. He also describes AI as a lens that magnifies human tendencies, and he locates the deeper problem in an appetite for untruth, with deepfakes and the 2024 elections raised as a concrete worry.

The later sections turn contentious. Suleyman is optimistic about efficiency gains in healthcare, education and agriculture, and about the lump of labor fallacy holding for the next couple of decades. He then argues that the real challenge is redistribution, because work is a means to well-being rather than an end. He backs the precautionary principle, meaning slower and more careful development even at the cost of forgone benefits. He also endorses the EU AI Act, and he disputes the "what about China?" framing, arguing that shared self-preservation interests can restrain a race to the bottom. The conversation also touches on low public trust in governments.

Sharpest takeaway

Mustafa Suleyman argues that AI will create radical abundance and solve major problems, but only if we resist a 'race to the bottom' on governance and values; containment of powerful AI models is fundamentally impossible as they proliferate into open-source and smaller forms, requiring proactive democratic institutions and international cooperation rather than competitive escalation.

  • AI computation is growing 10x annually with plans for 1000x more compute in 5 years, enabling planning across multiple time horizons and solving problems from agriculture to climate
  • The core containment problem is that AI algorithms are intellectual property expressible in a few sheets of paper; as models shrink to run on phones, bad actors gain access regardless of corporate safety measures
  • Geopolitical competition between US and China creates pressure toward a 'race to the bottom' where both sides abandon safety principles, but this is a self-fulfilling prophecy that can be resisted through appeal to shared preservation interests

The claims · ranked44 claims · weighted by value

This asset isn't compiled yet

You're seeing its claims, ranked. Compile it to build the argument threads, weight them, and check each claim against your library — the full view.

0.74

Three of the six co-founders of OpenAI passed through DeepMind either to give talks or as team members, including Ilya Sutskever (current chief scientist of OpenAI), who was an intern at DeepMind in 2011, indicating DeepMind's early leadership in deep learning and influence on subsequent AI institutions.

factualhigh valueestablishednovelty 1/4durability 4/4· Mustafa Suleyman

“the current chief scientists and co-founder of openai the creators of chat gbt was one of our interns back in 2011. Jeffrey Hinton who was the who subsequently became the um one of the heads of AI at Google and is known now as the Godfather of AI recently in the Press worried about the consequences he was our first advisor our paid advisor I think his salary was 25 000 pounds a year to advise us so I think three of the six co-founders of openai at some point passed through deepmind either to give talks or were actually members of the team”

0.70

AI will amplify both the very best and the very worst of humanity; the technology is fundamentally neutral and operates as a lens that magnifies existing human tendencies and capabilities.

causalhigh valueestablishednovelty 1/4durability 4/4· Mustafa Suleyman

“that's the challenge of AI that is a good lens for understanding the impact that AI has in general it is going to amplify the very best of us and it's also going to amplify the very worst of us”

0.68

Open-source AI models are increasingly powerful and only trailing closed models; over 10 years, as models become smaller, cheaper, and more transferable, open-source versions will eventually match closed models, meaning that even if major labs successfully contain dangerous capabilities, angry teenagers in garages will have access to equally powerful models without safety constraints.

forecasthigh valuecontestednovelty 2/4durability 3/4· Mustafa Suleyman

“if that trajectory continues over 10 years you get much much more powerful models that are much smaller and more you know transferable and you know people then who want to use them to cause harm have an easier time of it I think that's a really important distinction that there are you know the leading companies you Google deepmind you know open AI who have the biggest models now and they're a relatively small number of these ones and they are bigger and more powerful but not far behind are a whole bunch of Open Source ones and so the question is then for your containment can you prevent the open source ones which will potentially be available to the you know angry teenager in his garage or her garage can those ones be controlled or not”

0.68

There is an insatiable, addictive, dopamine-driven appetite for untruth that sells faster and spreads quicker than truth; this foundational psychological fact, not AI itself, is the root challenge, but AI and other technologies accelerate the spread of untruth.

causalhigh valuecontestednovelty 2/4durability 3/4· Mustafa Suleyman

“we have an appetite an insatiable addictive dopamine hitting appetite for untruth you know it sells quicker it spreads faster and that's a foundational question that we have to address I'm not sure that it's a new risk that AI imposes it's something that Ai and other Technologies accelerate”

0.64

Large language models have been trained on all data on the internet, which includes instructions for manufacturing biological and chemical weapons, representing a genuine dangerous capability that all major AI developers (Google, DeepMind, OpenAI) have observed and had to address

factualhigh valueestablishednovelty 1/4durability 2/4· Mustafa Suleyman

“there are certain aspects of these models which provide really good coaching on how to manufacture biological and chemical weapons it's one of the capabilities that all of us developing large language models over the last year have observed they've been trained on all of the data on the internet and much of that information contains potentially harmful things”

0.63

AI can optimize existing industrial systems for efficiency (e.g., cooling infrastructure), enabling us to produce more output with less resource consumption across healthcare, education, transportation, and agriculture

causalhigh valueestablishednovelty 0/4durability 3/4· Mustafa Suleyman

“we've seen many applications of AI where we're optimizing existing Industrial Systems we're taking the same big cooling infrastructure for example and when we making it much more efficient again we're doing more with less so in every area from Healthcare to education to Transportation we're very likely over the next two to three decades to see massive efficiencies”

0.62

The precautionary principle should guide AI development: we should go slower, be more careful, and potentially leave some benefits on the tree to avoid harms—a novel setup compared to past technology development where speed and benefit maximization were priorities

normativehigh valuecontestednovelty 1/4durability 3/4· Mustafa Suleyman

“we're being very clear clear that the precautionary principle is probably needed and that's a moment when we have to go a little bit slower be a little bit more careful and maybe leave some of the benefits on the tree for a moment before we pick that fruit in order to avoid harms”

0.62

Economic interdependence and appeal to shared interests in peace and prosperity between nations can align incentives, suggesting that US-China competition can be managed through mutual benefit rather than escalatory competition

causalhigh valuecontestednovelty 1/4durability 3/4· Mustafa Suleyman

“the more that we can appeal to that you know desire to you know have their citizens benefit from economic interdependence and from peace and prosperity and well-being we're both aligned in those incentives”

0.60

Geoffrey Hinton was DeepMind's first paid advisor at £25,000 per year in 2011 and is known as the 'Godfather of AI'; he subsequently became head of AI at Google and recently expressed concerns about AI consequences in the press

factualhigh valueestablishednovelty 0/4durability 4/4· Mustafa Suleyman

“Jeffrey Hinton who was the who subsequently became the um one of the heads of AI at Google and is known now as the Godfather of AI recently in the Press worried about the consequences he was our first advisor our paid advisor I think his salary was 25 000 pounds a year to advise us”

0.60

The deep learning revolution of the last 10 years enabled AI to interpret raw messy data—classifying images to identify dogs or cats, transcribing speech, translating languages—essentially teaching models to understand the content and meaning within data

factualhigh valueestablishednovelty 0/4durability 4/4· Mustafa Suleyman

“The Deep learning Revolution enabled us to make sense of raw messy data so we could use AIS to interpret the content of images classify whether an image contains dogs or cats what those pixels actually mean we can use it to understand speech so when you dictate into your phone and it transcribes it and Records perfect text we can use it to do language translation all of these are classification tasks we're essentially teaching the models to understand the messy complicated world of raw input data well enough to understand the objects inside that data”

0.57

Major AI companies (Google, DeepMind, OpenAI) can control and exclude dangerous capabilities from their models by removing harmful information from training data, since these companies operate within legal and ethical frameworks and don't want to cause harm

factualhigh valueestablishednovelty 0/4durability 2/4· Mustafa Suleyman

“that's a relatively easy thing to control and take out of the model at least when you're using a model that is manufactured by one of the big companies they want to abide by the law they don't want to cause harm so we basically exclude them from the training data and we prevent those capabilities”

0.57

Trust in governments and politicians is at historically low levels; in a Pew survey in the US, 35% of respondents think military rule would be a good thing, indicating society is in a fragile and anxious state vulnerable to AI-amplified problems

factualhigh valueestablishednovelty 0/4durability 2/4· Mustafa Suleyman

“the state we're in is is pretty bleak I mean trust in in governments and in politicians and the political processes as low as it has ever been um you know in in fact 35 percent of people interviewed in in a Pew study in the US think that Army rule would be a good thing so we're already in a very fragile and anxious State”

0.57

Framing geopolitical AI competition as a simple binary ('What about China?') with China having a maniacal, suicidal mission for global dominance at all costs is a dangerous oversimplification; in reality, China has self-preservation instincts like any nation

normativehigh valuecontestednovelty 1/4durability 2/4· Mustafa Suleyman

“I think that's a really Danger first it's sort of deep China as though China has this sort of like maniacal suicidal mission to at all costs you know sort of take over the world and you know be the next dominant Global power I mean so far I don't see any evidence of of that I mean you know I'm not ruling it out I'm not a you know sympathizer but I think we should just be wide-eyed about the actions they're actually taking at the moment they have a self-preservation Instinct just as we do”

0.57

What is required to manage AI risks is good functioning governance, oversight, and democratic institutions; companies are willing to expose themselves to audit, which represents a novel moment relative to past tech leaders who resisted oversight

normativehigh valuecontestednovelty 1/4durability 2/4· Mustafa Suleyman

“what's required is good functioning governance and oversight I mean the the companies are open and willing to expose themselves to audit and to oversight and I think that is a unique moment relative to past generations of tech CEOs and inventors and creators across the board”

0.56

The historical 'lump of labor fallacy' argues that automation doesn't eliminate jobs because efficiency creates new demand and new roles; we can point to roles today (app designer, prompt engineer) that couldn't have been conceived of 30 years ago

definitionhigh valuecontestednovelty 0/4durability 3/4· Mustafa Suleyman

“when we automate things and we make things more efficient we we create more time for people to invent new things and we create more health and wealth and that in itself creates more demand and then we we end up creating new goods and services to satisfy that demand and so we'll continually just keep creating new jobs and roles and you can see that in the last couple decades there are many many roles that couldn't even have been conceived of 30 years ago from App designer all the way through to the present day prompt engineer of a large language model”

0.56

Over the last 10 years, the amount of compute used for cutting-edge AI models has grown by 10x every single year—10x multiplied 10 times in a row—representing unprecedented exponential growth trajectory in technology history

factualhigh valueestablishednovelty 1/4durability 2/4· Mustafa Suleyman

“over the last 10 years every single year the amount of compute that we have used for The Cutting Edge AI models has grown by 10x so 10x 10x 10x 10x 10 times in a row now that is unprecedented in technology history nowhere else have we seen a trajectory anything like that”

0.55

The central challenge of AI abundance is redistribution: the goal should be to capture the value created by AI, convert it into economic value (rather than just GDP), and ensure it is redistributed so people are liberated from the obligation to work and can pursue well-being

normativehigh valuecontestednovelty 1/4durability 3/4· Mustafa Suleyman

“to me the goal of society is a quest for radical abundance how can we create more with radically less and liberate people from the obligation to work and that means that we have to figure out the question of redistribution and obviously that is an incredibly hard one and obviously I address it in the book but is that to focus on what does taxation look like in this new regime how do we capture the value that is created make sure that it's actually converted into Dollars rather than just a sort of value add to GDP”

0.53

Over the next two decades, it is highly unlikely we will see structural disemployment where people want to contribute labor but can't compete, suggesting the lump of labor fallacy likely holds true in the short to medium term

forecasthigh valuecontestednovelty 1/4durability 1/4· Mustafa Suleyman

“over the over the next two decades I think it's highly unlikely that we will see structural disemployment where people want to contribute their labor to the market and they just can't compete I think that's pretty unlikely there's certainly no evidence of it in the statistics today”

0.53

Deep fakes and synthetic media pose a threat to the 2024 elections; four billion people live in countries that will have elections next year, and concern about AI-driven election interference extends to 2028 and 2034

forecasthigh valueestablishednovelty 0/4durability 1/4· Unknown Speaker (Host)

“there are lots of concerns about you know deep fakes wrecking the 2024 elections four billion people live in countries that will have elections next year people are worrying about 2024 never mind 28 or 34”

0.52

AI algorithms are fundamentally ideas and intellectual property—they are pieces of know-how that can largely be expressed on three sheets of paper and are readily understandable to most people; implementation currently requires vast compute, but if that constraint is removed and AI can run on phones, the containment challenge becomes truly insurmountable

causalhigh valuespeaker onlynovelty 2/4durability 3/4· Mustafa Suleyman

“the darker side of my prediction is that these are fundamentally ideas you know their intellectual property it's knowledge and know-how an algorithm is something that can largely be expressed on three sheets of paper and actually is readily understandable to most people it's a little bit abstract but it you can wrap your head around it the implementation mechanism you know requires access to vast amounts of compute today but if in time you remove that constraint and you can actually run it on a phone which you ultimately will be able to do in a decade then that's where the containment challenge you know comes into view”

0.52

Within five years, AI will move from generating single outputs (perfect speech, video, images, text in isolation) to planning across multiple time horizons—requiring research, synthesis, negotiation, and multi-step execution like generating a new product by researching markets, designing, contacting manufacturers, negotiating, and marketing

forecasthigh valuecontestednovelty 2/4durability 1/4· Mustafa Suleyman

“we'll go from being able to perfectly generate so speech will be perfect video generation will be perfect image generation will be perfect language generation will be perfect to now being able to plan across multiple time Horizons so at the moment you could only say to a model give me you know a poem in the style of X give me a new image that matches these two Styles it's a sort of One-Shot prediction next you'll be able to say generate me a new product right in order to do that you would need to have that ai go off and do research to you know look at the market and see what was potentially going to sell what are people talking about at the moment it would then need to generate a new image of what that product might look like compared to other images so that it was different and unique it would then need to go and contact and manufacturer and say Here's the blueprint this is what I want you to make it might negotiate with that manufacturer to get the best possible price and then go and Market it and sell it those are the capabilities that are going to arrive you know approximately in the next five years”

0.52

The EU AI Act, which has been in draft for three and a half years, is thorough, robust, and sensible, providing a good model for AI governance that Suleyman endorses

normativehigh valuecontestednovelty 0/4durability 2/4· Mustafa Suleyman

“the EU AI act which has been in draft now for three and a half years is super thorough and very robust and pretty sensible and so in general I've been you know a fan of it and kind of endorsing it”

0.52

Rather than viewing work elimination as disaster, we should recognize that the goal of society is well-being, peace, and prosperity—not work itself; work is often an obligation people face out of necessity rather than choice, and should be viewed as means to those ends, not an end in itself

normativehigh valuespeaker onlynovelty 1/4durability 4/4· Mustafa Suleyman

“society and life and civilization is about well-being and peace and prosperity it's about creating more efficient ways to keep us productive and healthy many people you know probably in this room and including us enjoy our work we love our work and we're lucky enough and we're privileged enough to have the opportunity to do exactly the work that we want I think it's super important to remember that many many people don't have that luxury and many people do jobs that they would never do if they didn't have to work”

0.52

Intelligence has been the engine of creation throughout history, enabling us to solve problems and create abundance; AI should enable the same pattern of productivity gains that agriculture achieved (producing a kilo of grain with 2% of the labor required 100 years ago) across healthcare, climate, education, and transportation

causalhigh valuespeaker onlynovelty 1/4durability 4/4· Mustafa Suleyman

“intelligence has been the engine of creation everything that you see around you here is the product of us interacting with some environment to make a more efficient a more a cheaper table for example or a new iPad if you look back at history you know today we're able to create we're able to produce a kilo of grain with just two percent of the labor that was required to produce that same one kilo of grain 100 years ago so the trajectory of Technologies and scientific invention in general means that things are getting cheaper and easier to make and that means huge productivity gains right”

0.48

Philosophy is a systems-thinking tool that enables rigorous and clear thinking about complex problems, making it valuable for understanding technology and founding tech companies

normativehigh valuespeaker onlynovelty 1/4durability 3/4· Mustafa Suleyman

“I've always found philosophy as systems thinking tool it enables me to be rigorous and clear about what I think”

0.48

Suleyman became frustrated with the scale of impact achievable in non-profit work and realized that the profit incentive is one of the most important organizing principles for impact at scale, motivating his move toward technology companies

causalhigh valuespeaker onlynovelty 1/4durability 3/4· Mustafa Suleyman

“I soon got you know frustrated about the scale of impact um in our non-profit and I worked briefly for the mayor of London at the time Ken Livingston um as a human rights policy officer um and you know that was that was inspiring but I was also struggling with the scale of impact I I realized that you know if if I didn't capture what really makes us organized and effective as a species The Profit incentive then I was going to miss one of the most important things to happen in my lifetime”

0.48

AI's interpolation and generation capabilities will enable discovery of new knowledge, invention of new types of science, and novel solutions to problems by imagining things it has never seen before

forecasthigh valuespeaker onlynovelty 1/4durability 3/4· Mustafa Suleyman

“the AI is guessing the space between the dog the pink color and the yellow spots it's imagining something it's never seen before and that's exactly what we want from AI we want to discover new knowledge we want it to invent new types of science new solutions to problems and I think that's really what we're likely to get”

0.48

The best smartphone and laptop hardware in the world today is available to at least the top 2 billion people regardless of wealth, demonstrating that the most advanced technology can be democratized; AI should follow the same pattern

normativehigh valuespeaker onlynovelty 1/4durability 3/4· Mustafa Suleyman

“just as today no matter whether you are a you know a millionaire or you earn a regular salary we all get exactly the same access to the best smartphone and the best laptop that's an incredibly meritocratic story which we kind of have to internalize you know the Best Hardware in the world no matter how rich you are is available to at least the top 2 billion people and that is I think that is going to be the story that we see with respect to intelligence right”

0.48

If powerful tools are useful, they tend to get cheaper and spread far and wide; while proliferation has historically delivered immense benefits to society, the flip side is that powerful AI tools could empower bad actors to destabilize the world by advocating for their political, religious, and cultural agendas more effectively

causalhigh valuespeaker onlynovelty 1/4durability 3/4· Mustafa Suleyman

“if it's useful then it tends to get cheaper and therefore it spreads far and wide and in general so far that has delivered immense benefits to everybody in the world and it's something to be celebrated proliferation so far has been a really really good thing but the flip side is that if these are really powerful tools they could ultimately Empower a vast array of Bad actors to destabilize our world you know everybody has an agenda has a set of political beliefs religious beliefs cultural ideas and they're now going to have an easier time of advocating for it”

0.48

There are risks of centralization where powerful AI companies (including Suleyman's own company, Google, and other big tech providers) will accumulate significant power, and addressing open-source risks does not eliminate this centralization risk of super-powerful tech companies concentrating power relative to nation-states

causalhigh valuespeaker onlynovelty 1/4durability 3/4· Mustafa Suleyman

“I think that there are also risks of the centralization centralized question right this is clearly going to confer power on those who are building these models and running them you know my own company included Google and the other big Tech providers so we don't eliminate risk simply by addressing the open source Community we also have to figure out what the relationship is between these super powerful tech companies that have lots of resources and the nation state itself which is ultimately responsible for holding us accountable”

0.47

If AI development is managed correctly, we are headed towards an era of radical abundance where every person in the world has access to the best possible scientific advisor, research assistant, chief of staff, tutor, coach, and confidant—roles that today are exclusive to wealthy, educated people in peaceful societies

forecasthigh valuespeaker onlynovelty 2/4durability 3/4· Mustafa Suleyman

“I believe that if we can get that right we're headed towards an era of radical abundance imagine every great scientist every entrepreneur you know every person having the best possible Aid you know Scientific Advisor research assistant chief of staff tutor coach Confidant each of those roles that are today the you know exclusive Preserve of the wealthy and the educated and those of us who live in peaceful civilized societies those roles those capabilities that intelligence is going to be widely available to everybody in the world”

0.34

Mustafa Suleyman was a co-founder of DeepMind in 2010 and subsequently co-founded Inflection AI, establishing credibility as someone who has built successful AI companies and can offer a realistic assessment of AI's future beyond apocalyptic narratives

factualestablishednovelty 0/4durability 4/4· Unknown Speaker (Host)

“he was a co-founder of deepmind back in 2010. uh he then was a co-founder of inflection AI with Reed Hoffman”

0.34

Suleyman studied philosophy and theology at Oxford rather than computer science, bringing a systems-thinking approach to understanding AI and technology companies

factualestablishednovelty 0/4durability 4/4· Unknown Speaker (Host)

“Mustafa is actually not a computer geek you didn't study computer code right you studied philosophy and theology at Oxford”

0.34

Around 2007-2008, Facebook grew from zero to 100 million monthly active users in two years, demonstrating unprecedented speed of technology adoption that motivated Suleyman to learn about technology and seek partnerships to build tech companies

factualestablishednovelty 0/4durability 4/4· Mustafa Suleyman

“at the time I saw the rise of Facebook this was sort of around 2007 2008 and it had grown in the space of two years to 100 million monthly active users is and I was totally blown away at how quickly this was growing out of seemingly nowhere something completely new to me and so I set about on a quest to find anyone and everyone that would speak to me to teach me about technology”

0.34

Mustafa Suleyman met Demis Hassabis, his future DeepMind co-founder, through a school friend, when Hassabis was finishing his PhD in neuroscience at UCL, and together they founded DeepMind based on a bet on deep learning as a tool for artificial general intelligence

factualestablishednovelty 0/4durability 4/4· Mustafa Suleyman

“that's where I met my friend and co-founder of deepmind Mrs arbis because he was the brother of my best friend at the time from school and he was just finishing his PhD in neuroscience at UCL and we got together and you know the rest is history and at that time you had Patrol intelligence that was you know capable of replicating human intelligence or even succeeding it so just think this was 13 years ago the rest of us didn't even know this stuff was really going on”

0.34

DeepMind made a very big bet on deep learning at a time when most people were unaware that AI with the potential to replicate human intelligence was being developed; this timing was crucial to DeepMind's later success.

factualestablishednovelty 0/4durability 4/4· Mustafa Suleyman

“at the time you had Patrol intelligence that was you know capable of replicating human intelligence or even succeeding it so just think this was 13 years ago the rest of us didn't even know this stuff was really going on we made a very big bet on deep learning uh which is one of the primary tools that is powering this new Revolution um before anybody was involved in deep learning”

0.32

The book 'The Coming Wave' has received significant media attention (60+ appearances by Suleyman) because it addresses the hottest topic of the moment (AI) in a very interesting, thoughtful manner.

factualestablishednovelty 0/4durability 2/4· Moderator (Unknown)

“I believe you've heard you told me 60 appearances of various sorts so consider yourselves lucky you're 61 on this list but understandably the work the book has had a tremendous impact because it is very interesting very thoughtful and it's on the hottest Topic at the moment”

0.29

Suleyman dropped out of his Oxford philosophy degree at age 19 and co-founded Muslim Youth Helpline, a secular faith and culturally sensitive support service for young British Muslims, which he volunteered for for three years starting in 2003

factualspeaker onlynovelty 0/4durability 4/4· Mustafa Suleyman

“when I was 19 I actually dropped out of My Philosophy degree I didn't know that yeah I didn't finish and I was really motivated by the impact that I could have in the world I left to help start a charity at the time it was a telephone counseling service called Muslim youth helpline and it was a secular that I was a an atheist even though I had grown up with a Muslim background it was a secular service that was designed to provide faith and culturally sensitive support to Young British Muslims this was in 2 2003 and you know I I found myself at Oxford studying this very theoretical esoteric you know set of ideas and I wanted to put real things into practice in terms of my ethics and that was why I went to you know start the helpline and worked on that as a volunteer for three years”

0.29

Mustafa Suleyman is on the board of The Economist and has received a CBE (Commander of the Order of the British Empire) for visionary services and influence in the UK technology sector.

factualestablishednovelty 0/4durability 3/4· Moderator (Unknown)

“he was got a CBE a few years ago for his Visionary services and influence in the UK technology sector um he is also on the board of The Economist”

0.24

ChatGPT's release in November 2022 caused a collective global realization that artificial intelligence generative AI models in particular were about to change the world, reflected in widespread editorials, handwringing politicians, and endless focus on existential risks

factualestablishednovelty 0/4durability 2/4· Unknown Speaker (Host)

“chat DBT what came November last year [0:35] and it was only then that most people realized that artificial intelligence generative AI models in particular were about to change the world and suddenly there was a kind of collective Global oh my God this capability is extraordinary and it's been reflected in endless numbers of editorials handwringing politicians”

0.24

Suleyman's book makes a subtle argument that is not simply utopian optimism ('Becoming Nirvana') but rather addresses significant downsides and challenges of AI development alongside the potential upsides

definitionspeaker onlynovelty 0/4durability 4/4· Unknown Speaker (Host)

“but you didn't call your book you know becoming Nirvana you called it the coming wave and I'm told that you were thinking it the original title was going to be containment is not possible...but explain the argument you're making is not actually Nirvana is around the corner in fact it's a much much more subtle argument than that so tell us what the downsides are”

0.24

Suleyman started multiple unsuccessful businesses before DeepMind, including a technology company selling point-of-sale systems and Wi-Fi infrastructure in London restaurants around Notting Hill that was ahead of its time

factualspeaker onlynovelty 0/4durability 4/4· Mustafa Suleyman

“I had started a bunch of businesses before that two different businesses one actually a technology company selling electronic point of sale systems actually around here in Notting Hill uh in restaurants uh trying to put Wi-Fi infrastructure in there and so on that was a that was unsuccessful that was ahead of its time”

0.24

The book's original title was 'Containment Is Not Possible,' reflecting the argument that powerful AI tools will inevitably proliferate far and wide despite corporate safety measures, but this title was changed because it wouldn't have sold as well

factualspeaker onlynovelty 0/4durability 4/4· Unknown Speaker (Host)

“you didn't call your book you know becoming Nirvana you called it the coming wave and I'm told that you were thinking it the original title was going to be containment is not possible I'm glad you didn't call it that it wouldn't have sold so well”

0.20

Pi (Personal Intelligence) is Mustafa Suleyman's AI project at Inflection AI; it is designed to be a personal AI that acts as a chief of staff, prioritizing, planning, teaching, and supporting the user.

factualspeaker onlynovelty 0/4durability 3/4· Mustafa Suleyman

“Pi stands for personal intelligence and I believe that over the next few years everybody is going to have their own personal AI there are going to be hundreds of thousands of AIS in the world they'll represent businesses they'll represent Brands every government will have its own AI every non-profit every musician artist record label everything that is now represented by a website or an app is soon going to be represented by an interactive conversational intelligent service that represents the brand values and the ideas of whatever organization is out there and we believe that at the same time everybody will want their own personal AI one that is on your side in your corner helping you to be more organized helping you to make sense of the world it really is going to function as almost like a chief of staff or you know prioritizing planning teaching supporting supporting you”