Max Tegmark remains one of the most interesting scientists alive. His book, Life 3.0: Being Human in the Age of Artificial Intelligence (2017), is must reading for those wanting a quick and accessible survey of how artificial intelligence is changing, and will challenge, the way we live.
Tegmark was interviewed today on the Lex Fridman Podcast. For those of you who do not know Fridman, I say check him out. He is an MIT researcher in machine learning with a wide range of interests. Today marked his 153rd interview; in fact, the very first interview he did was also with Max Tegmark, himself a professor at MIT and a co-founder of the Future of Life Institute. You can count on Fridman to interview thought leaders and researchers in artificial intelligence, although his interviews often go well beyond that topic. The interviews are typically long and free-wheeling. Today's Tegmark interview is more than three hours long.
I don't listen to Fridman often enough. Candidly, I am so depressed by the state of the world and the sorry state of the news media, that I prefer to walk, as I do each day for a few hours, in silence rather than listen to more discordant noise. Black lives don't matter all that much to me, and I don't think the election was stolen. I am tired of hearing from partisan groupies.
But when I saw Fridman had Tegmark on, I was hooked. Tegmark is brilliant. I sometimes wonder if we engage in specious partisan bickering because AI is making us all superfluous. Are the overheated claims about identity the death rattle of folks who are being supplanted by machines?
I urge you to listen to the interview. It can provide a sense of optimism about the future.
Why?
Because at about the one hour mark in the interview -- I tell you this so that you need listen to the entire interview -- the conversation turns to one of the few things I've found exciting in the news in recent months. It is a new site that Tegmark created called Improve the News. You can find it by linking to www.improvethenews.org.
Tegmark is concerned about the fact that way we consume news these days increases polarization. Increasingly, folks rely on social media, rather than newspapers, for news. (I am as guilty of this as anyone; I gave up on The New York Times this past summer, once I saw its dedication to the 1619 Project and its determination to recast the entirely of our history into a tawdry narrative of racial grievances.) The result is that the algorithms that select what we see appeal to what motivates us to click onto a title. Clicks are, after all, what yields advertising revenue for the likes of Facebook, Twitter and Google. We're directed to silos based on the largely negative emotions social media algorithms detect.
Shouldn't well informed people read across the ideological spectrum? How do we break out of our silos? Indeed, how do we even know we're in a silo?
Enter the Improve the News project.
Tegmark and company have created an algorithm, a tool that directs a machine to select and sort data, along certain dimensions: political ideology, support for the established order, writing style, depth of coverage, recency, shelf life. You the reader have the opportunity my means to selecting among the variables to have news gathered to suit your preferences.
You're thinking -- how does this solve the "silo" problem and why won't this merely reinforce the problem? The answer is simple: If you engage with the system, you can switch from left to right on the same topic, getting a sample of what others are seeing. It's an eye-opening exercise and it's easy to use.
I can't vouch for the reliability of the algorithms. Indeed, I noticed certain anomalies. Infowars, a client of mine, is regarded by many as far right. Yet a story on a suspicious van in Tennessee appears in the "left" assortment of news under the topic of Crime and Justice. Similarly, the Daily Kos, regarded as left-leaning by most folks, is in the right wing assortment of stories about impeachment.
Perhaps that is the point. News isn't theology, or a doctrinal exercise -- at least it's not supposed to be. It may well be the algorithm detects less content than it does what will interest folks along the political spectrum.
I highly recommend that you spend some time on improvethenews.org. Play with the setting. See all the news that is fit to find its way to the Internet, rather than what an editor thinks is fit to print. Compare and contrast the left's and the right's treatment of the same topic. Set yourself free from the ideological silo Big Tech has placed you in.
I remain convinced that independent of the convictions, values and preferences that divide us, there is a world than can be known. Learning about that world and how to live in it with one another is imperative if we are to survive both individually and as a species. Nothing in the current political landscape I survey gives me optimism. I say a pox on both left and right, what I crave is reliability. Can AI set us free from ideology?
I will report further on Tegmark's bold experiment. I want to hear from you what you think of it, too.