Original article was published by Valentín Muro on Artificial Intelligence on Medium
How can we steer the boat in philosophy and make more philosophers worry about philosophical problems and not philosopher’s problems?
This is a difficult issue. It would be very easy to say “oh just do it, just go and sacrifice yourself, ruin your life, make sure you don’t have a future but do the right thing”. And I’m not gonna say that. The recommendations that I made in the past to my students or to people in an early stage of their career has always been that you need to do two things at the same time. You need to keep academia happy, be as if you were ok with the mechanism. So if they want you to do history of philosophy, you do history of philosophy, if they want you to do second order logic, that’s ok too. The thing is not to abandon that but to also do the other half. And the other half is to do what you think is important, what you really believe in. Now, the first half: the history, the logic, the not dealing with philosophical problems, has one advantage: it will give you a little bit of training. It’s like a gym. The question is: do I ever play? Do I actually go and play, say, football. I’m running on a treadmill all the time and maybe I’m very fast at running but I never practice, I never do philosophy. So, the question becomes not just I’ll stop doing that and do something else but do that and, at the same time, move into something that’s really important. Now, today I cannot believe what a huge need there is for thinkers, for people like you guys. Basically, at the right stage in their lives — I’m told I cannot call you young but it’s not a problem — at a different age. They don’t really have a legacy, in terms of old people who have always thought about the world in a certain way, basically a bit of virginity, mentally speaking.
We are exposed to an amazing society, that’s changing in front of our eyes at a pace unprecedented. With technologies we’ve never seen before. We are providing the foundations for the information societies of the future and it’s just extraordinary that potentialities, in terms of what philosophy can do, to see them go to a conference and see people talking about some minor Byzantine philosopher and did he really said that or not. It makes me cry because it is such a waste of energy, intelligence and opportunity. And also, we are not giving anything back to society. It’s crucial not to confuse this with applied philosophy, that’s not what I’m talking about. Applied philosophy is a joke. I’m talking about the Platos, the Aristotles, the Augustines, the Hobbes, the Descartes, Leibniz, these people who were talking to their time. I mean, pick up one of the classics top 25 philosophers in the world, and I bet you always find that they had a dialogue with tradition, they were talking with other philosophers, but above all they were talking of issues that were present at the time. Hobbes, John Locke, Bertrand Russell, Frege, I mean you name it. Descartes, Kant, Hegel, Rousseau. All of them, in their own tradition, they weren’t talking about that other philosopher that was saying this or that. They were dealing with the world first hand. So dealing with the world first hand to me means dealing with the information society, dealing with the technologies that are surrounding us, because they’re everywhere. So if you do bioethics, digital technology, if you do ontology and metaphysics, well, it’s changing in their eyes; do we understand really what the world is like?
Yesterday we had a long discussion in another meeting about the new ontology of things. Again, totally to be reinterpreted. So, whatever you touch is going to be something’s that going to do with technology, with digital, etc. Are we ready? Are we prepared? The University is not doing that.
Well, you know, it’s one of those moments where the old generation is coming to an end, the old paradigm and the new paradigm is coming in. Unless you guys start implementing the new paradigm, there will be a delay.
When it comes to misconceptions about AI and all that Skynet stuff, well, she mentioned that I started working as a sort of journalist for this newspaper. And what I try to do every week is to debunk some myths. For example, I wrote about the humans behind AI, many times in the other end typing and doing things like that. Do you think is important to dispel these myths and spend our time saying there’s no General Artificial Intelligence coming, and things like that?
I think it’s crucial for several reasons. First of all, because there are real problems. We saw some of them together at the meeting at the Society [SADAF]. There are real issues, so it’s not like we can waste our time debating some science fiction scenario, that is very funny or entertaining. No. We also mentioned some today. Today was more for the general public. But we are being constantly nudged and influenced in our decisions, more and more by all the technology — more or less smart — AI, that is surrounding us. So, there are really two problems at once. The first: bad science, bad philosophy, we shouldn’t be doing that. The other one is, on top of that, because we are doing all these stupid things, we are not doing the good stuff. So we are not taking care of understanding, discussing, making sure that we are going in the direction of where the real issues are. So, for example, take responsibility. Where do we allocate responsibility when it comes to a smart or AI system? How do we decide who is responsible for what? And do we want to delegate some decisions to AI systems or not? Take, for example, the discussion we are having in Europe on whether we have, as human beings, a right to appeal to decisions taken by machines, because is happening on a daily basis. When you go to a bank and there’s an algorithm that decides whether you get a mortgage or not. Is that okay? Shouldn’t it be not okay? Do I have a right to appeal? Of course, this becomes very practical, and a bit legal. But at the beginning of all this, there’s a philosophical understanding and decision. For example, in a kantian way, AI should be used to treat people always as ends, never as mere means. And therefore whatever AI you get to develop, keep this in mind, that’s crucial.
So, I would say two things: debunk all this rubbish that there is around, although there’s money to be made by not selling the rubbish, and concentrate on the real problems, because they are coming and they are huge. If we do both then mission accomplished.
I was reading your book the other day and there you mention that is really important for people to become digitally literate, that we need a digitally literate society. I always try to get people to learn how programming works, not so they can work as programmers but rather being able to understand the way software works. So what are your suggestions for getting to a digitally literate society?
I think that is a huge task that we have above our shoulders, to make sure that people don’t develop a sense of magic like, I don’t know, this is a black box that only the engineer, the logician, the AI expert, can understand. Not that everyone needs to know how a TV works or how a computer works. No, as long as people can relax and don’t think that there’s magic that goes through the TV, and it’s not magic in their box, and no magic in AI. For example, the debates today on the transparency and explainability of neural networks, which you probably discuss in your newspaper pieces, you hear a lot of people saying “not even the engineer can explain the decision”, not true! I mean, this is stuff that we built in a lab. You really think an engineer has no idea of what is happening, that it’s like a magic wand that you point and then something comes out? Of course not! But what the problem means is that we have the wrong sense of what it means to have an explanation. So in this particular example, normally in other contexts, imagine someone asking if I can explain on Monday morning, 8:30, there’s so much traffic in Oxford.
— I say, “well, the school’s are open, it’s rainy, I guess that’s why there’s so much traffic today, normally there isn’t.” Then someone comes and says “oh no, that’s not an explanation, I want to know why every car is there”. And so I reply “well, I don’t know”, and then the other one exclaims “it’s a mystery!” Of course it’s not a mystery! The schools are open at that time, the university is open at that time, it’s rainy, therefore the cars are there. Although you don’t have an explanation for every single car. So back to neural networks, you don’t know how every node contributes to the final decision in terms of threshold, up, and down, of course not. But that is not the point.
“So you don’t have an explanation, it’s a mystery!”. No, it isn’t. You give a thousand pictures of cats to an algorithm until it detects a cat.
I think we have imported the wrong kind of explanation, mostly from physics, where if I drop this pen on the floor, I will need to use Newtonian laws to explain why it’s falling. That’s a different world. So in complex systems the explanation needs to be at a different level of abstraction. It can’t be at the minute level where every element attracts any other element. Because ẁe know that already three elements mean the classic problem, the three body problem. We don’t have the equations to explain it.
So what I’m talking about here it’s bad philosophy of science and it’s bad computer science. So back to your point, we need to teach people what’s going on, we need to debunk what’s going on, debunk, once again, that all the mystery, all the magic and then people will relax about the technology. The impact? Now that’s where we need to work, that is crucial, because it’s changing the world as we speak.
Let me give you an example. Recently the local council decided to remove the coins to pay for the car park next to the railway station. And they implemented all in terms of mobile phones, smartphones, and credit card. And you may think it’s a great idea. It isn’t! It’s a university city, most students don’t have credit cards, old people don’t have smartphones. And all of a sudden people was like “okay, I can’t park the car, because I can’t go to the railway station, park the car and pay with coins”. And so now they have a mixed system. But it was bad strategy, bad understanding of what’s really needed, and a lot of damage is done by not thinking enough.
I assume you are worried about all this antiscience discourse that is becoming more and more prominent and is worrying many of us that come from philosophy and have an understanding of some of the issues at hand. Do you think we can put the blame not just on bad science but on some bad philosophy, such as the rise of relativism?
Absolutely. We have a lot of responsibility. Specially if you talk in terms of bad philosophy and not just Philosophy Departments but bad philosophical thinking in our society, like bad conceptual thinking. Then it becomes absolutely a huge responsibility of philosophers. We lower our guard thinking job done, everything’s going fine, end of history. Those were the 90s, there were lots of things that were going well: the wall in Berlin, the end of the South African apartheid, the world was finally going in the right train, finally we are going towards a good future. The intelligentsia lowered the guard. We became all relativistic and there are no facts, it’s all about ideology and you know what? The facts will bite you back. Because there were people who took advantage of this. We really made ourselves co-responsible for fake news, for “alternative facts”, for the Trump rubbish, or the Brexit rubbish. And if you keep going that way, and trust me: I have meetings where people, intellectuals, go “I don’t believe your facts”. Really? Okay, jump out of that window and it’s a fact you’re gonna crash.
So it’s a key responsibility on our side to be rational and evidence based, but don’t exaggerate because most of the world is complicated. There’s many more things than just choice theory, theorems and axioms. So the trouble is that if by every moment there’s rubbish around us, postmodernist crap basically, “it’s all interpretation, there are no facts, it’s all about the ideology”. Okay, whatever, because someone took advantage of that and screw up the world. But the reaction cannot be some form of rigid Enlightenment, some neo positivism or Carnapian view. That’s insane too. Logic is just this much in all this bit of life. There’s a lot of reasonable stuff, brash intelligence which is not covered by a couple of lines of code or calculus. It doesn’t mean that it falls out of interpretation. It’s just that there are so many intelligent, well-educated, evidence-based, fact-based ways of reasoning that we’ve been developing since Socrates. It’s called philosophy. So good philosophy, is that. Anything else, is a joke.
What I’m getting upset about it’s that it’s a joke that those people, the intelligentsia, they are taking advantage of that because they sell books, they go around and do some chit chat, so it’s easy. “Nothing is ever going to work, it’s all about power, power generates suffer”. Really? Then why are we here? Get out, I want to do something, I want to make the world a better place. And they go “oh, it’s all interpretation, there’s nothing in there, all is just a linguistic game”. Well, excuse me, because tomorrow we need to go in the street and explain to the guy who doesn’t have a job, or the lady who’s just been abused, or someone who has been raped that it’s all interpretation. I have zero patience with that, because we don’t have time to waste. This generation, the next and the game is over, in terms of the digital revolution, in terms of the environmental impact. We just don’t have time to indulge in “intellectual game playing”. Who is paying for this? It’s the rest of the world. And I can’t bear the view of a world that’s going on bad as it’s crashing just because a bunch of intellectuals have to play the “smart game”.