[HN Gopher] My Favorite Book on AI
       ___________________________________________________________________
        
       My Favorite Book on AI
        
       Author : f1shy
       Score  : 100 points
       Date   : 2025-01-05 16:28 UTC (1 days ago)
        
 (HTM) web link (www.gatesnotes.com)
 (TXT) w3m dump (www.gatesnotes.com)
        
       | thorum wrote:
       | > The historian Yuval Noah Harari has argued that humans should
       | figure out how to work together and establish trust before
       | developing advanced AI. In theory, I agree. If I had a magic
       | button that could slow this whole thing down for 30 or 40 years
       | while humanity figures out trust and common goals, I might press
       | it. But that button doesn't exist. These technologies will be
       | created regardless of what any individual or company does.
       | 
       | Is that true, though? Training runs for frontier models don't
       | happen without enormous resources and support. You don't run one
       | in your garage. It doesn't happen unless people make it happen.
       | 
       | Is this really a harder coordination problem than, say, stopping
       | climate change, which Gates does believe is worth trying?
        
         | liuliu wrote:
         | Genuinely asking, are we still trying to "stop" the climate
         | change any more or are we in the "easing the incoming pain"
         | phase already?
        
           | cyberpunk wrote:
           | We have a PhD student ecologist in the family and they've
           | lost all hope. It's quite depressing. Doesn't look like the
           | next round of world governments have it high on the agenda
           | also.
        
             | zombiwoof wrote:
             | Isn't it sad , that our government and companies quickly
             | solved AI energy problems by bootstrapping Nuclear again,
             | but in 30 years can't make a single unified decision on
             | anything else
        
               | TeMPOraL wrote:
               | It'll be ironic if AI solves climate change by just
               | existing, i.e. by giving the world a compelling reason to
               | stop hiding behind austerity and anti-growth bullshit
               | rhetoric, and to double down on energy production
               | instead, thus forcing us to tackle the "how do we do it
               | without cooking ourselves" problem up front, since it
               | cannot be deferred.
        
               | cyberpunk wrote:
               | It won't start moving until the boomer generation are
               | out, I think. Let's just hope we can do better when it's
               | our turn I guess.
        
           | beezlebroxxxxxx wrote:
           | Powerful, and wealthy in particular, people are more
           | interested in the latter than the former. When vast wealth is
           | built off of maintaining the status-quo there is very little
           | incentive for implementing changes that threaten the status-
           | quo "organically."
           | 
           | I get hope when I read this essay from Harpers [0]. But I
           | actually think it will be more like Paolo Bacigalupi's "The
           | Water Knife" [1].
           | 
           | [0]: https://harpers.org/archive/2021/06/prayer-for-a-just-
           | war-fi...
           | 
           | [1]: https://en.wikipedia.org/wiki/The_Water_Knife
        
           | basch wrote:
           | do you mean stop as in reverse the trend by limiting bad
           | behavior or do you mean stop like climate engineering. there
           | is always the option to use science to modify the climate on
           | purpose.
        
         | add-sub-mul-div wrote:
         | Sam Altman convincing the ultra-rich that AI is the next best
         | place to put their too much money after they've run out of
         | other places to put it is not anywhere near the spirit of what
         | he meant by people "working together." He's saying that while
         | in a dysfunctional society, AI will only magnify the
         | dysfunction.
         | 
         | Fighting climate change is a means towards the end of having a
         | livable environment, developing AI is a means towards the end
         | of having a better society. But, whereas fixing the environment
         | would be its own automatic benefit, having AGI would not
         | automatically improve the world. Something as seemingly
         | innocuous and positive as social networking made a lot of
         | things worse.
        
           | zombiwoof wrote:
           | This rings true. The internet exposed the dysfunctional
           | wealth inequality (look at all the billionaires it created) ,
           | social networks exposed the dysfunctional human
           | communications and manipulation (ie Jan 6 was a love fest) ,
           | and now AI will just take both of those to the next level.
           | Extreme wealth, extreme poverty and mass manipulation of
           | people
           | 
           | If anybody things AI will cure cancer or something grandiose
           | like that they are propping up their stock portfolio.
           | 
           | AI will be the harbinger of the last wave of human growth
           | before we all end up killing each other over the price of
           | eggs or whatever else AI regurgitation machine decides
        
         | lanthissa wrote:
         | Climate change is the byproduct of the desired outcome, energy.
         | Advanced AI, if you buy Yuval's argument is the threat in and
         | of itself.
         | 
         | So Climate Change is a problem that can be 'solved' while the
         | main goal is pursued. This is ideologically consistent with
         | gates investment in terrapower. Where as AI isn't because the
         | desired outcome is the threat not a bi-product.
         | 
         | So your questions a bit flawed fundamentally.
         | 
         | As for gate's point is it true, almost certainly yes, the game
         | theory is peruse and lie that you aren't, or openly pursue. You
         | can't ever not pursue because you do not and cannot have
         | perfect information.
         | 
         | Imagine how much visibility china would demand from the US to
         | trust it was doing nothing, far more than they could give, and
         | vice versa.
         | 
         | Do you think the us is going to give its adversaries tracking
         | and production information its most advanced chips? It would
         | never, and if they did why would other powers trust it if
         | theres every reason to lie.
        
           | joe_the_user wrote:
           | _Climate change is the byproduct of the desired outcome,
           | energy. Advanced AI, if you buy Yuval 's argument is the
           | threat in and of itself._
           | 
           | That's just a matter of how you slice your concepts. You
           | could say burning oil is a threat in and off itself, for
           | example. Or oppositely, "the threat of bad AI" is a byproduct
           | of "useful AI".
           | 
           |  _So Climate Change is a problem that can be 'solved' while
           | the main goal is pursued._
           | 
           | I don't think many people trying to solve climate change are
           | trying to end industrial society. They are trying to find an
           | energy source that doesn't produce Co2 pollution.
        
         | tim333 wrote:
         | AI research is global and of strategic value with both the US
         | and China competing. I don't see one stopping research while
         | the other cracks ahead. Similar problems exist with curbing co2
         | which hasn't gone very well to date.
        
       | the_arun wrote:
       | With all due respect - I would have hoped to see a list of other
       | AI books reviewed with a recommendation. Currently the article
       | seems like a Preface for the book - "The Coming Wave".
        
       | davideg wrote:
       | I remember people on HN having a less than favorable sentiment
       | about Mustafa Suleyman:
       | 
       | https://news.ycombinator.com/item?id=39757330
       | 
       | Regardless, he's certainly been in the right places to understand
       | AI trends and Gates' write-up makes it sound like an intriguing
       | distillation. Thanks for posting!
        
       | Barrin92 wrote:
       | It's an okay book but there isn't really anything in it that you
       | couldn't infer after you've read the first 10%. A lot of common
       | sense warnings about risks from AI, bioweaopns, cyberattacks etc
       | but it's all very generic. There's no chapter in it that I found
       | had any genuine insight. An interesting chapter would have been
       | "what if I'm completely wrong and all we get is a bunch of meme
       | generators and the next bubble", but that never appears to be a
       | possibility.
       | 
       | It's oddly enough the case with a lot of books that end up on
       | Gates recommended lists. I saw someone recently say, maybe a bit
       | too mean, that we might make it to AGI because Yuval Noah Harari
       | keeps writing books that more and more look like they're written
       | by ChatGPT and it's not entirely untrue for a lot of the stuff
       | Gates recommends.
        
       | option wrote:
       | the author of that book is a _non-technical_ co-founder of
       | DeepMind who currently leads Microsoft AI efforts.
        
         | add-sub-mul-div wrote:
         | Apologies if I've missed your point, but if what you're hinting
         | at is that we should take his ideas about the social impact of
         | AI less seriously because he's not _deep into writing Rust
         | code_ all day, that 's just laughable.
        
           | asdasdsddd wrote:
           | Well yes, we don't really know what this thing is yet, and
           | the only ones who kind of understand it are the researchers
           | themselves.
        
           | option wrote:
           | no, my point is that before reading a book it is often
           | helpful to know who the author is and what is their track
           | record
        
       | thangalin wrote:
       | An alpha reader for my hard sci-fi novel wrote:
       | 
       | "Hey just wanted to let you know I started autonoma - only
       | started - I'm at page 45 now but I'm really digging this - love
       | getting into a story about AI and the image of the scene in Japan
       | in the game - super great - and the scene with the coy wolves-
       | I'm totally in."
       | 
       | The novel has a take on AGI and ASI that diverges from our fear
       | of machines that will destroy/control/enslave humanity. I'd be
       | grateful for any other alpha readers who'd like to give me their
       | thoughts on the story, especially with respect to the economic
       | ramifications. See my profile for contact details.
        
       | Insanity wrote:
       | The author is the head of Microsoft AI. Gates might not be
       | entirely unbiased here :)
       | 
       | edit: as a semi-related question for folks here. How often do you
       | 'vet' authors of non-fiction books prior to reading the book?
        
         | NilMostChill wrote:
         | depends on the subject matter.
         | 
         | If it's something i have no grounding in then understanding the
         | authors' potential biases is useful.
         | 
         | If it's something I'm relatively familiar with or close enough
         | that i think I'll be able to understand the application of
         | potential biases in realtime then i don't usually bother.
         | 
         | This issue is sometimes somewhat alleviated by reading multiple
         | sources for the same/similar information.
         | 
         | YMMV
        
         | ladyprestor wrote:
         | I always do a little bit of research on the author online
         | beforehand. If I'm going to read a non-fiction book I need to
         | know if the author is credible.
         | 
         | I do believe Bill Gates might be a little bit biased here. I
         | read the book some months ago, and while I can't say it's a bad
         | book, I wouldn't call it a favorite either.
        
         | zeofig wrote:
         | >The author is the head of Microsoft AI. I'm disappointed but
         | not surprised.
        
         | ccppurcell wrote:
         | Funny you should say that because I just read his Wikipedia
         | page and a couple of articles about him. He and Gates are
         | successful salesmen and managers who dabbled in coding when
         | they were young. I don't expect any insight from them about the
         | effect of technologies on society or anything like that. The
         | idea that they are intellectuals or scholars is laughable.
        
         | pjmorris wrote:
         | > How often do you 'vet' authors of non-fiction books prior to
         | reading the book?
         | 
         | I misread the question as 'How' rather than 'How often', but
         | I'll repeat Jerry Weinberg's heuristic. He'd wait until three
         | people he trusted recommended a book before reading it, as a
         | way to filter for quality. He used it as a way to manage his
         | limited time ("24 hours, maybe 60 good years" - Jimmy Buffett),
         | but it also works to weed out books not worth mentioning.
         | 
         | In terms of 'how often', pretty often.
        
         | senko wrote:
         | I read all the negative/neutral GoodReads comments of the book
         | (and author's other books if I'm not familiar with the author,
         | and maybe Wikipedia if I want to dig deeper).
         | 
         | 99% of books I learn about from recommendations (HN, blogs,
         | other books), and the pattern I see is that the
         | source/recommender are usually at the similar "popsci" level.
         | 
         | I sometimes get it wrong. In most cases I just waste a few
         | hours. The worst mistake was taking Why We Sleep to heart
         | before I read the rebuttal. I still think it's fine, but more
         | on a Gladwell level.
         | 
         | Im Suleyman's case, I recognize the name from Inflection
         | shenanigans, so already have a bias against the book to start
         | with.
        
       | WillAdams wrote:
       | Interesting critical review at:
       | 
       | https://www.goodreads.com/book/show/90590134-the-coming-wave
       | 
       | >...
       | 
       | >Given that The coming wave assumes that technology comes in
       | waves and these waves are driven by the insiders, the solution it
       | proposes is containment--governments should determine (via
       | regulation) who gets to develop the technology, and what uses
       | they should put the technology to. The assumption seems to be
       | that governments can control access to natural choke points in
       | the technology. One figure the book offers is how around 80% of
       | the sand used in semiconductors comes from a single mine--control
       | the mine and you control much of that aspect of the industry.
       | This is not true though. Nuclear containment, for example, relies
       | more on peer pressure between nation states, than regulation per
       | se. It's quite possible to build a reactor or bomb in your
       | backyard. The more you scale up these efforts, the more it's
       | likely that the international community will notice and press you
       | to stop. Squeezing on one of these choke points it more likely to
       | move the activity somewhere else, then enable you to control it.
       | 
       | >...
       | 
       | >At its heart this is a book by and insider arguing that someone
       | is going to develop this world-changing technology, and it should
       | be them.
        
       | bambax wrote:
       | > _I've always been an optimist, and reading The Coming Wave
       | hasn't changed that._
       | 
       | I'm not an optimist, but I fail to see the dangers of AI. I think
       | it's more likely we will be wiped out by nuclear war, or climate
       | change, or the collapse of biodiversity and ecosystems that
       | result in worldwide famines, before AI is advanced enough to
       | constitute any kind of threat to our existence.
        
         | randcraw wrote:
         | The dangers are not in AI, per se. Like nuclear fission and
         | fusion, the danger is in how the technology 1) may be misused
         | by corporations that are clueless to or disinterested in the
         | damage it can inflict, and 2) surely will be dis-regulated by
         | the increasingly stupid and malignant boobs infecting
         | Washington.
        
       | Bjorkbat wrote:
       | Can't take this seriously knowing that this is the same Mustafa
       | Suleyman who...
       | 
       | - Was basically acqui-hired by Microsoft from Pi AI (seems a
       | little biased to recommend a book from one of your own)
       | 
       | - Left DeepMind due to allegations of bullying
       | (https://en.wikipedia.org/wiki/Mustafa_Suleyman#DeepMind_and_...)
       | 
       | - Allegedly yelled at OpenAI employees because they weren't
       | sharing technologies frequently enough
       | (https://www.nytimes.com/2024/10/17/technology/microsoft-open...)
       | 
       | But what do I know, maybe if I read it and regurgitate its
       | contents in a not-too-obvious way I can get an AI policy job.
        
         | tim333 wrote:
         | Suleyman seems to have got ahead in AI basically by being mates
         | with Demis Hassabis and joining him in the company founding. He
         | doesn't seem to have achieved much in actual AI, more things
         | like setting up a "Muslim Youth Helpline" and being a "policy
         | officer on human rights for Ken Livingstone."
        
       | zombiwoof wrote:
       | How can anyone take Mustafa Suleyman seriously
        
       | HPMOR wrote:
       | This book is middling at best. From a literary perspective it's
       | terrible. It reads like GPT-3.5 wrote it. Now from an
       | introduction to a new idea perspective and understanding how AI
       | will affect society, its __fine__. The book is full of
       | contradictions. Suleyman regularly points out how we've __never__
       | successfully constrained a disrupting technological innovation
       | and then says we NEED to here. I mean, absolutely absurd stuff.
       | Ostensibly ends on an optimistic note, but is actually much more
       | nihilistic.
        
         | mjfl wrote:
         | The begging for regulatory capture in the AI business is so
         | egregious that I think the government should nationalize all
         | the AI companies 'to keep us safe' but really to prevent these
         | shysters from making money.
        
         | Sverigevader wrote:
         | Hmm, didn't we do this for cloning? I remember hearing about
         | this on Lex Fridman when he interviewed Max Tegmark.
         | 
         | If I recall correctly, the entire world is in agreement that
         | cloning is illegal, and even that some people in China (could
         | be just one) even went to prison for it.
        
           | LPisGood wrote:
           | If you could use cloning to make lots of money more labs
           | would be doing it.
        
             | AlexCoventry wrote:
             | You probably _could_ make lots of money from human cloning,
             | if it weren 't illegal?
        
               | WalterBright wrote:
               | If I could clone myself, I'd make a few bug fixes.
        
         | WalterBright wrote:
         | > It reads like GPT-3.5 wrote it
         | 
         | These days it would be surprising if an author didn't generate
         | at least some of the text with AI, or direct an AI to improve
         | the prose.
        
       | amelius wrote:
       | From the guy who didn't see the internet coming ...
        
         | thomassmith65 wrote:
         | Gates wrote his 'internet tidal wave' email mid 1995
         | (https://wired.com/2010/05/0526bill-gates-internet-memo/) which
         | was only two years after Berners-Lee publicly released Mosaic.
         | 
         | By the late 1990s, Microsoft's competition (including Netscape
         | and Apple) were nearly dead. In fact, the browser that Apple
         | originally shipped with OS X was M$ Internet Explorer.
         | 
         | Gates was several months late to the web, but it's not like he
         | missed the boat.
        
           | wrs wrote:
           | I went from Apple to Microsoft in 1995 partially because
           | Microsoft was so far ahead of Apple on the Internet. At the
           | time, Apple was entirely concerned with promoting eWorld.
           | (Ironic because Apple had a /8 IP allocation and was
           | processing something like a quarter of Usenet traffic well
           | before this.) They both had to get out of their walled garden
           | "compete with AOL" models, but MSFT did it faster.
        
             | thomassmith65 wrote:
             | The other problem with the Mac in those years was that
             | there was no decent web browser.
             | 
             | Windows MSIE eventually surpassed the usability,
             | functionality and popularity of Netscape, but Microsoft's
             | Mac version of MSIE did not.
             | 
             | In the late 1990s, many websites did not render or function
             | correctly on Macintosh.
        
           | fsckboy wrote:
           | absolutely everybody in software was talking about the
           | internet in 1995, and for the most part already on the
           | internet, and companies were already pivoting to the internet
           | left and right. It made me realize how discretionary is all
           | the work we do in these large industries, because whatever we
           | were working on in 1993 and 1994 no longer mattered, we were
           | now working on enabling whatever assets we had to be
           | compatible with the internet.
        
       | saneshark wrote:
       | I just finished listening to it on audible. It is certainly
       | thought provoking, but full of contradictions as others have
       | mentioned. Namely that this technology cannot be contained, and
       | yet that it must be contained is pretty doom and gloom. The
       | prognostications about artificial intelligence are hardly as
       | scary as the ones made around genetic sequencing -- that you can
       | buy a device for 30k that will print pathogens and viruses for
       | you out of your garage. That's some scary stuff.
        
         | downrightmike wrote:
         | You can buy plasmids and make whatever bacteria you want for a
         | few decades now. AI may help, but it certainly doesn't cost
         | $30k to cause mischief.Pretty sure I learned that in Bio 102
        
           | Balgair wrote:
           | I just want to echo this here and in a bit different wording:
           | AI will provide step-by-step guides on how to make viruses
           | that just about any idiot can follow, for very cheap, and in
           | under a year time frame.
           | 
           | I really _really_ hope I 'm missing something big here.
        
             | mdorazio wrote:
             | Having a step-by-step guide and _actually being able to
             | follow it_ are two very different things. If you follow
             | YouTube channels like The Thought Emporium you 'll see how
             | hard it is just to duplicate existing lab results from
             | published sources in biology. To go a step further and
             | create new dangerous things without also getting yourself
             | killed in the process is a pretty tall order.
        
               | Keyframe wrote:
               | _Having a step-by-step guide and actually being able to
               | follow it are two very different things._
               | 
               | exactly. we'll see how far it goes. it might be a more
               | elaborate draw the rest of an owl guide, like:
               | 
               | 1. obtain uranium-238
               | 
               | 2. fire up the centrifuge for isotope separation
               | 
               | 3. drop yellowcake into it
               | 
               | 3. collect uranium-235
               | 
               | ...
        
               | downrightmike wrote:
               | You missed the part where you turn uranium metal into a
               | gas for the centrifuge to work in the first place
        
               | energy123 wrote:
               | We should be talking about the more abstract problem of
               | asymmetric defense and offense.
               | 
               | Imagine that nukes were easy to make with household
               | items. That would be a scenario where offense is easy but
               | defense is hard. And we would not exist as a species
               | anymore.
               | 
               | Once a hypothetical technology like this is discovered
               | one time, it's not possible to put the genie back in the
               | bottle without extreme levels of surveillance.
               | 
               | We got lucky that nukes were hard to make. We had no idea
               | that would be the case before nuclear physics was
               | discovered, but we played Russian Roulette and survived.
        
         | root_axis wrote:
         | They said the same thing about the anarchist's cookbook 30
         | years ago.
        
       | mjfl wrote:
       | I wish he would read 'The Israel lobby' by John Mearsheimer and
       | Stephen Walt. Gates has some responsibility as an elite
       | billionaire to safegaurd our democracy.
        
       | jack_pp wrote:
       | https://www.fimfiction.net/story/62074/friendship-is-optimal
        
       | netfortius wrote:
       | Some folks really need to read Thibault Prevost's "Les prophetes
       | de l'IA", in order to comprehend the motivation behind work like
       | Suleyman's.
        
       | PunchTornado wrote:
       | this is from Mustafa Suleyman, don't bother opening the link.
       | 
       | If people don't know him, he is the classic impostor who just
       | goes by contributing nothing to the field, but investing big in
       | pr and bots.
        
         | mistrial9 wrote:
         | news flash - he got a promotion, staff and budgets since then
        
           | randcraw wrote:
           | After he was demoted at DeepMind for poor performance as a
           | manager, that is. Suleyman is no technologist. He's a
           | marketing promoter. He attended one year of college before
           | dropping out to help sell DeepMind.
           | 
           | You have to wonder what's going on in Gates' head these days
           | to not recognize the lack of substance in such a book, and in
           | its author.
           | 
           | Far better books on the possible futures of modern AI are
           | Stuart Russell's "Human Compatible" or Brian Christian's "The
           | Alignment Problem", both of which predate the boom in LLMs
           | but still anticipate their Achilles heel -- the inability to
           | control what they learn or how they will use it.
        
       | tim333 wrote:
       | TED talk by the author, probably with similar content to the book
       | https://youtu.be/KKNCiRWd_j0
        
       | figers wrote:
       | In an early chapter he talks about how well LLMs knowing medical
       | and legal information but doesn't mention how it makes things
       | up... Was hoping he'd discuss the challenges and hurdles right
       | away...
        
       | DwnVoteHoneyPot wrote:
       | For those negative on the book, do you have a better suggestion
       | for me to read? Maybe I should just ask ChatGPT.
        
       | greenthrow wrote:
       | Bill Gates is not some great thinker. He was born on third base
       | and in the right place at the right time, and then absolutely
       | ruthless once Microsoft had power in the industry. As a thought
       | leader he is extremely mediocre.
        
       | pockmarked19 wrote:
       | My favorite book on AI is Sutton's "reinforcement learning".
       | Looking just at the url I knew this would be some pop-sci tripe
       | but leaving this comment here in case people want something other
       | than what they can tout on twitter/X.
        
       | gerash wrote:
       | Is he recommending the book because MS hired Suleyman?
        
       | jasoneckert wrote:
       | While I've enjoyed the small bursts of wisdom from many of Bill
       | Gates shorter talks, I haven't found anything noteworthy in his
       | written reviews and books. His viewpoints are often bizarre and
       | radical. I still chuckle when remembering reading his conviction
       | that ISDN would become the dominant Internet technology before
       | the year 2000 in his book "The Road Ahead" (1995). It even seemed
       | bizarre back then to me.
        
       | talles wrote:
       | This reads like a PR written blog post recommending a PR written
       | book
        
       ___________________________________________________________________
       (page generated 2025-01-06 23:00 UTC)