I work with a bunch of mathematicians, philosophers and computer scientists, and we sit around and think about the future of machine intelligence, among other things. Some people think that some of these things are sort of science fiction-y, far out there, crazy. But I like to say, okay, let's look at the modern human condition. (Laughter) This is the normal way for things to be.
Radim s mnogo matematičara, filozofa i računalnih znanstvenika s kojima razmišljam o budućnosti strojne inteligencije, među drugim stvarima. Neki ljudi misle da su te stvari u zoni znanstvene fantastike, nešto ludo, nestvarno. Ali ja volim reći ok, pogledajmo uvjete modernoga života. (Smijeh) Ovo je normalno stanje stvari.
But if we think about it, we are actually recently arrived guests on this planet, the human species. Think about if Earth was created one year ago, the human species, then, would be 10 minutes old. The industrial era started two seconds ago. Another way to look at this is to think of world GDP over the last 10,000 years, I've actually taken the trouble to plot this for you in a graph. It looks like this. (Laughter) It's a curious shape for a normal condition. I sure wouldn't want to sit on it. (Laughter)
Ali ako malo razmislimo o tome, mi smo zapravo tek nedavno stigli na ovaj planet, mi, ljudska vrsta. Kad bismo gledali na Zemlju kao da je stvorena prije godinu dana, onda bi ljudska vrsta bila stara deset minuta. Industrijska je era počela prije dvije sekunde. Drugi pogled na ovo jest taj da promatramo svjetski BDP u zadnjih 10 000 godina. Odvojio sam nešto vremena i to za vas grafički prikazao. Izgleda ovako. (Smijeh) Neobičan je to oblik za uobičajene uvjete. Ja sigurno ne bih sjedio na njemu. (Smijeh)
Let's ask ourselves, what is the cause of this current anomaly? Some people would say it's technology. Now it's true, technology has accumulated through human history, and right now, technology advances extremely rapidly -- that is the proximate cause, that's why we are currently so very productive. But I like to think back further to the ultimate cause.
Zapitajmo se koji je uzrok ove trenutne anomalije. Neki bi ljudi rekli da je tehnologija. Istina je da se tijekom povijesti tehnologija nagomilavala i u ovom se trenutku tehnologija jako brzo razvija -- to je neposredni uzrok, i zbog toga smo trenutno jako produktivni. Ali razmislimo o krajnjem uzroku.
Look at these two highly distinguished gentlemen: We have Kanzi -- he's mastered 200 lexical tokens, an incredible feat. And Ed Witten unleashed the second superstring revolution. If we look under the hood, this is what we find: basically the same thing. One is a little larger, it maybe also has a few tricks in the exact way it's wired. These invisible differences cannot be too complicated, however, because there have only been 250,000 generations since our last common ancestor. We know that complicated mechanisms take a long time to evolve. So a bunch of relatively minor changes take us from Kanzi to Witten, from broken-off tree branches to intercontinental ballistic missiles.
Pogledajte ova dva vrlo istaknuta gospodina: imamo Kanzija -- koji je svladao 200 leksičkih znakova što je nevjerojatan podvig. I Ed Witten koji je započeo drugu revoluciju superstruna. Ako pomno pogledamo, uočit ćemo ovo: u principu istu stvar. Jedna je malo veća, možda ima par specifičnosti u načinu na koji je sklopljena. Međutim, ove nevidljive razlike ne mogu biti jako složene jer nas dijeli samo 250 000 generacija od zadnjeg zajedničkog pretka. Znamo da se složeni mehanizmi dugo razvijaju. Tako da nas hrpa relativno malih promjena vodi od Kanzija do Wittena, od polomljenih grana do interkontinentalnih balističkih raketa.
So this then seems pretty obvious that everything we've achieved, and everything we care about, depends crucially on some relatively minor changes that made the human mind. And the corollary, of course, is that any further changes that could significantly change the substrate of thinking could have potentially enormous consequences.
Onda je poprilično očito da sve što smo postigli i sve do čega nam je stalo uvelike ovisi o relativno malim promjenama koje čine ljudski um. Nužna je posljedica ta da sve daljnje izmjene koje bi mogle značajno promijeniti način razmišljanja mogu imati potencijalno ogromne posljedice.
Some of my colleagues think we're on the verge of something that could cause a profound change in that substrate, and that is machine superintelligence. Artificial intelligence used to be about putting commands in a box. You would have human programmers that would painstakingly handcraft knowledge items. You build up these expert systems, and they were kind of useful for some purposes, but they were very brittle, you couldn't scale them. Basically, you got out only what you put in. But since then, a paradigm shift has taken place in the field of artificial intelligence.
Neki od mojih kolega smatraju da smo na rubu nečega što bi moglo izazvati takve temeljite promjene, a radi se o strojnoj superinteligenciji. Prije je umjetna inteligencija bila pisanje naredbi u kutiju. Ljudski bi programeri mukotrpno ručno stvarali stavke znanja. Izgradite takve stručne sustave koji su za određene svrhe korisni, ali su bili krhki i nisu bili mjerljivi. U principu ste dobivali samo ono što ste unijeli. Ali, od onda se dogodila promjena u pristupu u polju umjetne inteligencije.
Today, the action is really around machine learning. So rather than handcrafting knowledge representations and features, we create algorithms that learn, often from raw perceptual data. Basically the same thing that the human infant does. The result is A.I. that is not limited to one domain -- the same system can learn to translate between any pairs of languages, or learn to play any computer game on the Atari console. Now of course, A.I. is still nowhere near having the same powerful, cross-domain ability to learn and plan as a human being has. The cortex still has some algorithmic tricks that we don't yet know how to match in machines.
Danas se uglavnom radi o učenju strojeva. Umjesto ručnog unosa prikaza znanja i značajki, stvaramo algoritme koji uče, i to često iz čistih perceptualnih podataka. Načelno je to ista stvar koju čini ljudsko dijete. Rezultat je taj da UI nije ograničena na jedno područje -- isti sustav može naučiti prevoditi bilo koju kombinaciju jezika ili naučiti igrati bilo koju računalnu igru na Atariju. Naravno, UI nije ni blizu posjedovanja iste jake, unakrsne sposobnosti učenja i planiranja raznih područja kao što to radi ljudsko biće. Korteks ima neke algoritamske trikove koje ne znamo kako prenijeti strojevima.
So the question is, how far are we from being able to match those tricks? A couple of years ago, we did a survey of some of the world's leading A.I. experts, to see what they think, and one of the questions we asked was, "By which year do you think there is a 50 percent probability that we will have achieved human-level machine intelligence?" We defined human-level here as the ability to perform almost any job at least as well as an adult human, so real human-level, not just within some limited domain. And the median answer was 2040 or 2050, depending on precisely which group of experts we asked. Now, it could happen much, much later, or sooner, the truth is nobody really knows.
Tako da je pitanje koliko smo blizu tome da uskladimo te trikove. Prije nekoliko godina proveli smo istraživanje s vodećim stručnjacima za UI kako bismo vidjeli što misle, i jedno od pitanja bilo je "Što mislite, do koje će godine biti 50% vjerojatnosti da ćemo postići strojnu inteligenciju na razini ljudske inteligencije?" Ljudsku smo razinu definirali kao sposobnost da se bilo koji zadatak obavi jednako dobro kao što bi ga obavio odrastao čovjek, znači, stvarna ljudska razina, a ne samo unutar određenog područja. Srednji je odgovor bio 2040. ili 2050., ovisno o skupini stručnjaka kojoj smo pitanje postavili. To bi se moglo dogoditi puno, puno kasnije ili ranije, istina je ta da nitko to zapravo ne zna.
What we do know is that the ultimate limit to information processing in a machine substrate lies far outside the limits in biological tissue. This comes down to physics. A biological neuron fires, maybe, at 200 hertz, 200 times a second. But even a present-day transistor operates at the Gigahertz. Neurons propagate slowly in axons, 100 meters per second, tops. But in computers, signals can travel at the speed of light. There are also size limitations, like a human brain has to fit inside a cranium, but a computer can be the size of a warehouse or larger. So the potential for superintelligence lies dormant in matter, much like the power of the atom lay dormant throughout human history, patiently waiting there until 1945. In this century, scientists may learn to awaken the power of artificial intelligence. And I think we might then see an intelligence explosion.
Ali ono što znamo jest to da krajnja granica procesuiranja informacija u strojevima uvelike nadilazi ograničenja biološkog tkiva. Radi se o fizici. Biološki se neuron ispaljuje na možda 200 herca, 200 puta u sekundi. Ali čak i današnji tranzistor radi u gigahercima. Neuroni se polako šire u aksone, najviše 100 metara u sekundi. Ali u računalima signali mogu putovati brzinom svjetlosti. I tamo postoje ograničenja u veličini, isto kao što ljudski mozak mora stati u lubanju, ali računalo može biti veličine skladišta, ili čak veće. Potencijal superinteligencije prikriven je i čeka, slično kao što je snaga atoma bila prikrivena tijekom povijesti, strpljivo čekajući 1945. godinu. U ovom su stoljeću znanstvenici možda otkrili kako probuditi moć umjetne inteligencije. A onda bismo mogli svjedočiti eksploziji inteligencije.
Now most people, when they think about what is smart and what is dumb, I think have in mind a picture roughly like this. So at one end we have the village idiot, and then far over at the other side we have Ed Witten, or Albert Einstein, or whoever your favorite guru is. But I think that from the point of view of artificial intelligence, the true picture is actually probably more like this: AI starts out at this point here, at zero intelligence, and then, after many, many years of really hard work, maybe eventually we get to mouse-level artificial intelligence, something that can navigate cluttered environments as well as a mouse can. And then, after many, many more years of really hard work, lots of investment, maybe eventually we get to chimpanzee-level artificial intelligence. And then, after even more years of really, really hard work, we get to village idiot artificial intelligence. And a few moments later, we are beyond Ed Witten. The train doesn't stop at Humanville Station. It's likely, rather, to swoosh right by.
Većina ljudi, kada razmišlja o tome što je pametno, a što glupo, mislim da otprilike zamišlja ovako nešto. Na jednom kraju imamo seosku budalu, a na drugome Eda Wittena ili Alberta Einsteina, ili već preferiranog gurua. Ali iz perspektive umjetne inteligencije, ta slika vjerojatno izgleda ovako: sve počinje na točki nulte inteligencije, a onda nakon puno, puno godina napornog rada možda konačno dođemo do umjetne inteligencije na razini miša, nešto što se može kretati kroz pretrpan prostor jednako dobro kao miševi. A onda nakon još puno, puno godina napornog rada i mnogo ulaganja, možda dođemo do umjetne inteligencije na razini čimpanze. A onda, nakon još puno godina vrlo, vrlo napornog rada, dođemo do seoske budale umjetne inteligencije. I nakon nekoliko trenutaka nadiđemo Eda Wittena. Tu napredak ne prestaje već se ubrzano nastavlja.
Now this has profound implications, particularly when it comes to questions of power. For example, chimpanzees are strong -- pound for pound, a chimpanzee is about twice as strong as a fit human male. And yet, the fate of Kanzi and his pals depends a lot more on what we humans do than on what the chimpanzees do themselves. Once there is superintelligence, the fate of humanity may depend on what the superintelligence does. Think about it: Machine intelligence is the last invention that humanity will ever need to make. Machines will then be better at inventing than we are, and they'll be doing so on digital timescales. What this means is basically a telescoping of the future. Think of all the crazy technologies that you could have imagined maybe humans could have developed in the fullness of time: cures for aging, space colonization, self-replicating nanobots or uploading of minds into computers, all kinds of science fiction-y stuff that's nevertheless consistent with the laws of physics. All of this superintelligence could develop, and possibly quite rapidly.
To ima duboke implikacije, posebno kad se radi o pitanju moći. Na primjer, čimpanze su snažne -- dvostruko su jače od muškaraca. A ipak, sudbina Kanzija i njegovih prijatelja više ovisi o tome što mi ljudi radimo nego o tome što same čimpanze rade. Jednom kada se stvori superinteligencija, sudbina će čovječanstva možda ovisiti o tome što ona radi. Razmislite. Strojna inteligencija je zadnji izum koji će čovječanstvo ikada trebati. Strojevi će onda biti bolji u izumljivanju od nas, i oni će to raditi u digitalnom vremenskom rasponu. Načelno to znači teleskopiranje budućnosti. Zamislite sve moguće lude tehnologije koje bi ljudi mogli razviti: lijek protiv starenja, kolonizacija svemira, samoreplicirajući nanoboti, unošenje uma u računalo, svakakve znanstveno-fantastične stvari koje su u skladu sa zakonima fizike. Superinteligencija mogla bi se razviti, i to dosta brzo.
Now, a superintelligence with such technological maturity would be extremely powerful, and at least in some scenarios, it would be able to get what it wants. We would then have a future that would be shaped by the preferences of this A.I. Now a good question is, what are those preferences? Here it gets trickier. To make any headway with this, we must first of all avoid anthropomorphizing. And this is ironic because every newspaper article about the future of A.I. has a picture of this: So I think what we need to do is to conceive of the issue more abstractly, not in terms of vivid Hollywood scenarios.
Superinteligencija s takvom tehnološkom zrelošću bila bi iznimno moćna, i bar bi u nekim slučajevima, mogla dobiti što želi. Onda bismo u budućnosti imali svijet oblikovan prema preferencijama te UI. Dobro je pitanje koje su te preference. Tu se stvari kompliciraju. Da bismo napredovali s tim, moramo prvo izbjeći antropomorfizaciju. A to je ironično jer svi novinski članci o budućnosti UI imaju sliku ovoga: Tako da mislim da trebamo pojmiti ovo pitanje malo apstraktnije, a ne u obliku holivudskih scenarija.
We need to think of intelligence as an optimization process, a process that steers the future into a particular set of configurations. A superintelligence is a really strong optimization process. It's extremely good at using available means to achieve a state in which its goal is realized. This means that there is no necessary connection between being highly intelligent in this sense, and having an objective that we humans would find worthwhile or meaningful.
Inteligenciju trebamo promatrati kao proces optimizacije, proces koji usmjerava budućnost u određeni set konfiguracija. Superinteligencija je zaista jak proces optimizacije. Nevjerojatno je dobra u korištenju raspoloživih sredstava kako bi postigla svoj cilj. To znači da nužno ne postoji povezanost između visoke inteligencije u tom smislu i cilja koji bismo mi ljudi smatrali vrijednim ili smislenim.
Suppose we give an A.I. the goal to make humans smile. When the A.I. is weak, it performs useful or amusing actions that cause its user to smile. When the A.I. becomes superintelligent, it realizes that there is a more effective way to achieve this goal: take control of the world and stick electrodes into the facial muscles of humans to cause constant, beaming grins. Another example, suppose we give A.I. the goal to solve a difficult mathematical problem. When the A.I. becomes superintelligent, it realizes that the most effective way to get the solution to this problem is by transforming the planet into a giant computer, so as to increase its thinking capacity. And notice that this gives the A.I.s an instrumental reason to do things to us that we might not approve of. Human beings in this model are threats, we could prevent the mathematical problem from being solved.
Recimo da UI damo cilj da nasmiju ljude. Kad je UI slaba, izvodi korisne ili zabavne radnje zbog kojih se njihov korisnik smije. Kada UI postane superinteligentna, uvidjet će da postoji učinkovitiji način na koji postići taj cilj: preuzimanje kontrole nad svijetom i smještanje elektroda u mišiće lica ljudi kako bi izazvala neprestan, vedar osmijeh. Još jedan primjer, recimo da zadamo UI da riješi zahtjevan matematički problem. Kad UI postane superinteligentna, uviđa da je najučinkovitiji način za rješavanje toga problema preobrazba planeta u ogromno računalo, da bi povećala njegovu sposobnost razmišljanja. I imajte na umu da to daje UI instrumentalni razlog da nam radi stvari koje mi možda ne odobravamo. U ovom su primjeri ljudi prijetnja, mogli bismo spriječiti rješavanje matematičkih problema.
Of course, perceivably things won't go wrong in these particular ways; these are cartoon examples. But the general point here is important: if you create a really powerful optimization process to maximize for objective x, you better make sure that your definition of x incorporates everything you care about. This is a lesson that's also taught in many a myth. King Midas wishes that everything he touches be turned into gold. He touches his daughter, she turns into gold. He touches his food, it turns into gold. This could become practically relevant, not just as a metaphor for greed, but as an illustration of what happens if you create a powerful optimization process and give it misconceived or poorly specified goals.
Naravno, stvari neće baš tako poći po krivu; ovo su primjeri iz crtića. Ali važna je poanta: ako stvorite zaista moćan proces optimizacije kako biste maksimizirali cilj x, morate biti sigurni da vaša definicija x-a uključuje sve do čega vam je stalo. Ovo je lekcija na koju nas upozoravaju mnogi mitovi. Kralj Mida želi da se sve što dotakne pretvori u zlato. Dodirne kćer i ona se pretvori u zlato. Dodirne hranu, ona postane zlato. Ovo bi moglo postati i praktički važno, ne samo kao metafora za pohlepu, nego i kao prikaz onoga što se dogodi ako stvorite moćan proces optimizacije i date mu pogrešne ili loše određene ciljeve.
Now you might say, if a computer starts sticking electrodes into people's faces, we'd just shut it off. A, this is not necessarily so easy to do if we've grown dependent on the system -- like, where is the off switch to the Internet? B, why haven't the chimpanzees flicked the off switch to humanity, or the Neanderthals? They certainly had reasons. We have an off switch, for example, right here. (Choking) The reason is that we are an intelligent adversary; we can anticipate threats and plan around them. But so could a superintelligent agent, and it would be much better at that than we are. The point is, we should not be confident that we have this under control here.
Mogli biste reći da, ako računalo počne stavljati elektrode u lica ljudi, jednostavno ćemo ga isključiti. Pod a, to nije tako jednostavno ako smo već ovisni o sustavu -- na primjer, kako isključiti internet? Pod b, zašto čimpanze nisu isključile prekidač ljudske vrste, oni ili neandertalci? Zasigurno su postojali razlozi za to. Mi imamo prekidač, recimo, upravo ovdje. (Guši se) Razlog je taj što smo mi inteligentni protivnici; predviđamo prijetnje i planiramo kako ih izbjeći. Ali to bi mogao raditi i superinteligentni izvršitelj i to puno bolje nego što mi to radimo. Poanta je ovoga da ne smijemo biti uvjereni da sve držimo pod kontrolom.
And we could try to make our job a little bit easier by, say, putting the A.I. in a box, like a secure software environment, a virtual reality simulation from which it cannot escape. But how confident can we be that the A.I. couldn't find a bug. Given that merely human hackers find bugs all the time, I'd say, probably not very confident. So we disconnect the ethernet cable to create an air gap, but again, like merely human hackers routinely transgress air gaps using social engineering. Right now, as I speak, I'm sure there is some employee out there somewhere who has been talked into handing out her account details by somebody claiming to be from the I.T. department.
Mogli bismo si olakšati posao tako što bismo smjestili UI u kutiju, nekakvo sigurno softversko okruženje, virtualnu simulaciju stvarnosti iz koje ne može pobjeći. Ali kako možemo biti sigurni da UI neće pronaći rupu? Ako uobzirimo da hakeri neprestano pronalaze rupe, vjerojatno ne možemo biti sigurni u to. Isključimo Ethernet kabel da bismo stvorili zračni prostor, ali opet, ljudski hakeri rutinski nadilaze te prostore socijalnim inženjeringom. Dok ovo govorim, siguran sam da su negdje neku zaposlenicu nagovorili da otkrije detalje svog računa nekome tko je tvrdio da je iz IT odjela.
More creative scenarios are also possible, like if you're the A.I., you can imagine wiggling electrodes around in your internal circuitry to create radio waves that you can use to communicate. Or maybe you could pretend to malfunction, and then when the programmers open you up to see what went wrong with you, they look at the source code -- Bam! -- the manipulation can take place. Or it could output the blueprint to a really nifty technology, and when we implement it, it has some surreptitious side effect that the A.I. had planned. The point here is that we should not be confident in our ability to keep a superintelligent genie locked up in its bottle forever. Sooner or later, it will out.
Mogući su i kreativniji scenariji, na primjer, ako ste UI, možete zamisliti micanje elektroda u unutrašnjem sklopu kako biste stvorili radio valove kojima možete komunicirati. Ili se možete prevarati da ste neispravni i kada vas programeri otvore da vide što nije u redu s vama, pogledaju izvorni kod i -- Bum! -- manipulacija je počela. Ili bi moglo dati nacrt zaista divne tehnologije, a kada ga primijenimo, ima neke skrivene posljedice koje je UI isplanirala. Poanta jest ta da ne bismo trebali biti presigurni u svoju sposobnost da superinteligentnog duha možemo zauvijek čuvati u boci. Prije ili kasnije će se osloboditi.
I believe that the answer here is to figure out how to create superintelligent A.I. such that even if -- when -- it escapes, it is still safe because it is fundamentally on our side because it shares our values. I see no way around this difficult problem.
Vjerujem da je rješenje u pronalasku načina kako stvoriti superinteligentnu UI koja bi čak - kad - se oslobodi još uvijek bila sigurna jer je u osnovi na našoj strani jer dijeli naša uvjerenja. Ne vidim kako drukčije riješiti ovaj veliki problem.
Now, I'm actually fairly optimistic that this problem can be solved. We wouldn't have to write down a long list of everything we care about, or worse yet, spell it out in some computer language like C++ or Python, that would be a task beyond hopeless. Instead, we would create an A.I. that uses its intelligence to learn what we value, and its motivation system is constructed in such a way that it is motivated to pursue our values or to perform actions that it predicts we would approve of. We would thus leverage its intelligence as much as possible to solve the problem of value-loading.
Zapravo sam ja i optimističan kada mislim da je problem rješiv. Ne bismo morali pisati duge liste svega do čega nam je stalo, ili, još gore, unijeti to u neki računalni jezik kao C++ ili Python, to bi bio i više nego beznadan pothvat! Umjesto toga, stvorimo UI koja koristi svoju inteligenciju kako bi naučila što nam je važno, i čiji je motivacijski sustav konstruiran tako da bude motivirana za provođenje naših uvjerenja ili izvođenje radnji za koje predviđa da ih odobravamo. Tako bismo iskoristili njezinu inteligenciju što je više moguće kako bismo riješili ovaj problem.
This can happen, and the outcome could be very good for humanity. But it doesn't happen automatically. The initial conditions for the intelligence explosion might need to be set up in just the right way if we are to have a controlled detonation. The values that the A.I. has need to match ours, not just in the familiar context, like where we can easily check how the A.I. behaves, but also in all novel contexts that the A.I. might encounter in the indefinite future.
To se može dogoditi, a ishod bi mogao biti jako dobar za čovječanstvo. Ali to se ne događa automatski. Polazni uvjeti za eksploziju inteligencije morat će biti postavljeni na točno određen način ako želimo kontroliranu detonaciju. Vrijednosti UI moraju odgovarati našima, ne samo u poznatim kontekstima u kojima lako provjerimo kako se UI ponaša, nego i u novim kontekstima s kojima bi se UI mogla susresti u neodređenoj budućnosti.
And there are also some esoteric issues that would need to be solved, sorted out: the exact details of its decision theory, how to deal with logical uncertainty and so forth. So the technical problems that need to be solved to make this work look quite difficult -- not as difficult as making a superintelligent A.I., but fairly difficult. Here is the worry: Making superintelligent A.I. is a really hard challenge. Making superintelligent A.I. that is safe involves some additional challenge on top of that. The risk is that if somebody figures out how to crack the first challenge without also having cracked the additional challenge of ensuring perfect safety.
Postoje i neka ezoterična pitanja koja bi se morala riješiti: točni detalji teorije odlučivanja, kako se nositi s logičkom nesigurnosti i tako dalje. Tehnički problemi koji se moraju riješiti kako bi ovo uspjelo poprilično su veliki -- ne toliko veliki kao stvaranje superinteligentne UI, ali prilično veliki. Ovo je razlog za brigu: Stvaranje superinteligentne UI težak je izazov. Stvaranje sigurne superinteligentne UI dodaje još izazova. Rizik leži u otkrivanju kako savladati prvi izazov, ali bez shvaćanja kako savladati dodatni izazov, osiguravanje sigurnosti.
So I think that we should work out a solution to the control problem in advance, so that we have it available by the time it is needed. Now it might be that we cannot solve the entire control problem in advance because maybe some elements can only be put in place once you know the details of the architecture where it will be implemented. But the more of the control problem that we solve in advance, the better the odds that the transition to the machine intelligence era will go well.
Smatram da bismo trebali riješiti pitanje kontrole unaprijed kako bismo bili spremni kada za to dođe vrijeme. Može se dogoditi da ne možemo taj problem riješiti unaprijed jer se mogu posložiti samo neki elementi kada točno saznamo gdje će se provoditi. Što više problema po pitanju kontrole riješimo unaprijed, bolji su izgledi da će prijelaz u eru strojne inteligencije proći dobro.
This to me looks like a thing that is well worth doing and I can imagine that if things turn out okay, that people a million years from now look back at this century and it might well be that they say that the one thing we did that really mattered was to get this thing right.
Meni se čini da se to isplati raditi i mogu zamisliti da, ako sve prođe u redu, će ljudi za milijun godina gledati na ovo stoljeće i reći da smo napravili jednu stvar koja je stvarno važna, a ta je da smo ovo napravili kako treba.
Thank you.
Hvala.
(Applause)
(Pljesak)