Max Tegmark - Science and the Future of Humanity

We are privileged to be living when science is making astounding progress. Humanity has learned so much, so quickly. What will science achieve in another hundred years? A thousand years? A million years? Unimaginable? But what are the risks? Are they existential? For the future of humanity, what will science bring?




Max Tegmark is Professor of Physics at Massachusetts Institute of Technology. He holds a BS in Physics and a BA in Economics from the Royal Institute of Technology in Sweden. He also earned a MA and PhD in physics from University of California, Berkeley.


Closer to Truth, hosted by Robert Lawrence Kuhn and directed by Peter Getzels, presents the world’s greatest thinkers exploring humanity’s deepest questions. Discover fundamental issues of existence. Engage new and diverse ways of thinking. Appreciate intense debates. Share your own opinions. Seek your own answers.

[fullyawakened]

Although this was posted only an hour ago, Max is addressing already outdated concerns which probably dates this footage. He talks about his friends not being sure if an AI can write books or movies in 100 years from now but if you've been keeping up with the latest breakthroughs we are already well beyond that point. In fact, the issue is compounded here because it's not a single AI we have to worry about but rather the teams of AIs that we are weaving together to create these next gen networks. Separate AIs that have been developed for chat, driving, voice and face recognition, language, environment scanning, search engine, gaming, 3d simulation and 2 dozen others are linked together to allow something like the new driving AIs that let you give your vehicle novel instructions like "drive down the block and turn left by the flatbed truck, head til you see a dirt road and take a right just after the brown barn, go to you see the first grove of trees and drop the package on a red porch which should be on one of the first couple of houses on the road." All these neural networks that have been trained to excellence on one specific task are now integrated into devices that we have right now that can interpret those instructions and correctly execute them. Writing a book or movie is child's play for these things. They have been doing that at a level that passes the Turing test for almost 5 years already.

[bretnetherton]

Awareness is known by awareness alone.

[jeancharlesdecastledubajac]

When had this interview took place. Because some questions are now already answered. That shows how fast the A. I. Technology progresses.

[mtshasta]

scientists, atheists, agnostics and religious/theological people, should all agree on one thing--that improving our existence and helping other humans live better, regardless of their position or argument should be CenterPoint for all--anything less is anti-human and diabolical.

[paddlefar]

The thing I try to remember is that while modern advances in technology can be incredibly helpful, especially in regards to health benefits, knowledge, ease of life, etc, if used appropriately, there can be a dark side too, to be aware of. Most people have figured this out probably, that one has to be savvy and knowledgeable, or at least wise enough to know when some bit of technology is no longer being helpful in their lives and especially know when it is actually hurting them, and therefore needs to be used with caution or perhaps even abandoned. Social media come to mind here, because it is driven in many cases, heavily by any money that it can extract from the user by means of manipulation - to get a person to buy stuff they don’t really need. The more alone, or fearful or inadequate the advertisers and media manipulators can make people feel, the more likely, it’s statistically proven, that it works to gets those people to buy the products that are advertised, in the hopes that it will somehow make the buyer’s life somehow better, feel safer, make them look better, be more popular or make them think it will make them happier. It’s often totally bogus, and there are a lot of psychological tricks used, but the advertisers don’t care how harmful or manipulative it is, as long as it brings them more money.

[kricketflyd]

A discussion about the Phoenix cycle or predictions of the next cataclysm would be more realistic. People are naturally interested in documented events that are going to return timed by our celestial calendar in the stars.

[Stegosaurus]

I don't know what's going to happen.

[Self-Duality]

Science requires metaphysics for positively-oriented breakthroughs.

[ififif]

1:40 Dude, the sociology that's developed around nuclear weapons is one where we're playing a perpetual game of Russian roulette. Eventually we're going to lose that game. Avoiding a nuclear armageddon is a much more important existential topic of discussion than AI.

[philipm]

Few have explored these questions as inventively as Arthur C. Clark in Childhood's End.

[pathcoinfirst]

See "Forbidden Planet". The 1950's sci fi addressed the fundamental issue.

[jazzunit]

It needs to be aimed at understanding the natural world and life

[fluxophile]

Funny timing, with that EXPO article that just dropped.

[im-Anarchy]

some specific videos actually need longer videos

[LuciFeric]

There is no "line" between life and non life. Is salt alive? When i eat salt does the salt magically come to life? How about water and oxygen?

[danielteran]

Max looks exactly like he was back 20 yeas a go or more in his early videos!.

[jwbflyer]

Ok he missed his opportunity but I will tell you RLK here are the BIG question-

Who benefits. As this age is brought in, and how do we restructure our society and commerce and who will benefit because I can assure you with the utmost confidence there are power structures at work to reduce who actually benefits.

[MrJasonworkman]

I believe the question is how do we control the feelings of robots.

[rc]

5:05 that future is already here... today AI is generating context that was thought of as impossible just a few months earlier... everything now is changing by the day and I don't think we'll be able to catch up with all the progress within this year...

[B.S...]

Lets hope A.I. development is better managed than the Castle Bravo test. I don’t think a hostile autonomous super brain will give us a second chance. Rule #1 - teach it the golden rule and the Hippocratic oath.