philosophy

Artificial Intelligence

Artificial Intelligence is the attempt to construct machines that can think as well, or perhaps even better, than we can. A.I. started up around the 1950s, and progress was slow.

For example, playing chess is something that computers are very good at, but it took decades of effort to construct a machine that could defeat the worlds chess champion. In 1997, Deep Blue, constructed by IBM, beat Garry Kasparov, who at the time was the world champion.

But, for various reasons, progress in A.I. is rapidly accelerating. One of these reasons is the development of powerful “machine learning algorithms”.

Recently, a machine was built at Carnegie Mellon University (which, along with MIT, is a major hub of AI in the U.S.) using these machine learning algorithms. The machine plays poker, and demolished a handful of professional poker players. There are machines that play Go as well. And there are machines that do other things aside from playing games.

There could be a time when these machines are curing diseases and solving the world’s problems.

But some smart people (including Stephen Hawking and Elon Musk) are worried about these machines. What if they decide to destroy us?

Do you think AI is possible? Should we worry about these machines?

The Singularity

The Singularity is a hypothetical moment in human history at which we manage to construct a “superintelligence,” a machine that is smarter than us.

We are used to being the smartest thing on the planet so this will be a big change. Some claim that the singularity will be one of the most important moments in human history and it will transform everything.

Some also claim that the singularity is close. They think it will happen during your lifetimes.

And what happens if this machine that is smarter than us builds something that is smarter than it. And then this new machine builds something smarter than it, and so on. We might get to a point where the gap in intelligence between us and these machines are huge.

There are a number of ways AIs might become dangerous.

-Governments might weaponize them on purpose.

-They could see us like we are ants and not worth keeping around since they might be so much smarter than us.

-Or they could “accidentally” kill us.

Imagine that we tell a super intelligent machine to eliminate human suffering. Being very logical, the machine concludes that the easiest way to do that is to kill all of us (human can’t suffer if there are no humans).

Nick Bostrom is a philosopher that worries about these possibilities. Here’s a quote from him:

“Suppose we have an AI whose only goal is to make as many paper clips as possible. The AI will realize quickly that it would be much better if there were no humans because humans might decide to switch it off. Because if humans do so, there would be fewer paper clips. Also, human bodies contain a lot of atoms that could be made into paper clips. The future that the AI would be trying to gear towards would be one in which there were a lot of paper clips but no humans.”

What do we make of any of this? Worried yet?

Technological Unemployment

Let’s suppose that machines don’t kill us.

There is still another worry: they are going to take all our jobs.

We are already seeing this start to happen.

We are already starting to see self-driving cars in some places; they are going to take a lot of jobs (trucking etc).

Machines might even take highly skilled jobs from us.

What are we going to do when we see unemployment rates that are three times that of during the Great Depression?

Some have started to argue that what will be needed is a Universal Basic Income. Governments are going to have to give everyone a set amount of money.

Should we worry about this? What should we do? Is a universal basic income the solution?

The Simulation argument

In 2003, Nick Bostrom, a philosophy professor, wrote a paper called, “Are you living in a computer simulation?”

We now have a lot of computing power. One thing that we have been doing with this computing power is running various computer simulations of things. People simulate the weather on computers for example. Sporting events and so on.

But what if we decided to run a simulation of the history of our universe? Or maybe a lot of simulations of the history of our universe? We don’t have the computing power yet, but maybe we will one day.

Think about the future of humanity. There seems to be only a few possibilities.

1. "The fraction of human-level civilizations that reach a posthuman stage (that is, one capable of running high-fidelity ancestor simulations) is very close to zero", 

Or maybe we survive, but decide that we are not interested in simulating our universe. That is,

2. "The fraction of posthuman civilizations that are interested in running ancestor-simulations is very close to zero", or

maybe we survive, and maybe we decide to run a bunch of simulations of our universe. But if so, then there will be one reality, but in that reality they will be running very many simulations of reality. There will be lots of simulations of reality but only one reality, so it is more likely that we are in a simulation of reality than reality itself. That is,

3. "The fraction of all people with our kind of experiences that are living in a simulation is very close to one"

So, either we all die, we continue to progress but decide not to simulate the universe on computers, or we are almost certainly in a simulation ourselves. We are basically like characters in a video game.

“Many works of science fiction as well as some forecasts by serious technologists and futurologists predict that enormous amounts of computing power will be available in the future. Let us suppose for a moment that these predictions are correct. One thing that later generations might do with their super-powerful computers is run detailed simulations of their forebears or of people like their forebears. Because their computers would be so powerful, they could run a great many such simulations. Suppose that these simulated people are conscious (as they would be if the simulations were sufficiently fine-grained and if a certain quite widely accepted position in the philosophy of mind is correct). Then it could be the case that the vast majority of minds like ours do not belong to the original race but rather to people simulated by the advanced descendants of an original race. It is then possible to argue that, if this were the case, we would be rational to think that we are likely among the simulated minds rather than among the original biological ones. Therefore, if we don't think that we are currently living in a computer simulation, we are not entitled to believe that we will have descendants who will run lots of such simulations of their forebears.”

— Nick Bostrom, Are you living in a computer simulation?, 2003

Are we in a simulation?

The Turing Test

The roots of A.I. are often said to be in the work of Alan Turing, a British mathematician and logician.

Turing was an interesting guy. He published a paper in 1936 (when he was 22 years old) called “On Computable Numbers.” In it, he developed the theoretical foundation of computer science.

During World War II, he cracked the Nazi’s enigma code, the code the German military used to send messages back and forth. This was a huge advantage for the Allies.

Turing died in the early 1950s in tragic circumstances.

But right before he died, he published a paper called “Computing Machinery and Intelligence.” This is the paper credited with starting A.I.

The most famous idea in the paper is the Turing test (though Turing called it the “Imitation Game.”)

It is a test to determine if a machine is thinking or not. No computer has ever passed it, though sometimes someone will claim that one has.

“Turing (1950) describes the following kind of game. Suppose that we have a person, a machine, and an interrogator. The interrogator is in a room separated from the other person and the machine. The object of the game is for the interrogator to determine which of the other two is the person, and which is the machine…The interrogator is allowed to put questions to the person and the machine of the following kind: “Will X please tell me whether X plays chess?” Whichever of the machine and the other person is X must answer questions that are addressed to X. The object of the machine is to try to cause the interrogator to mistakenly conclude that the machine is the other person; the object of the other person is to try to help the interrogator to correctly identify the machine.”

Turing says if a machine can fool the questioner 70% of the time into thinking it is a person, then the machine thinks.

What do we make of this? Could a machine that passes the Turing test be made? Is this a good test of intelligence?

Chinese room argument.

An objection to AI (and the Turing test) made by John Searle, a philosopher at Berkeley.

“Imagine a native English speaker who knows no Chinese locked in a room full of boxes of Chinese symbols (a data base) together with a book of instructions for manipulating the symbols (the program). Imagine that people outside the room send in other Chinese symbols which, unknown to the person in the room, are questions in Chinese (the input). And imagine that by following the instructions in the program the man in the room is able to pass out Chinese symbols which are correct answers to the questions (the output). The program enables the person in the room to pass the Turing Test for understanding Chinese but he does not understand a word of Chinese.

Searle goes on to say, “The point of the argument is this: if the man in the room does not understand Chinese on the basis of implementing the appropriate program for understanding Chinese then neither does any other digital computer solely on that basis because no computer, qua computer, has anything the man does not have.”

Thoughts?