This article is part of the On Tech newsletter. You can Login here to get it on weekdays. We’ll be back for the rest of the week and on Monday.
The idea of computers taking over people was a staple of science fiction. My colleague Cade Metz says we might get there. Perhaps.
Cade talked to me about a technology called GPT-3 that it is learning Write that it looks like it was composed by people. This technology could help accelerate the development of computer programs that can hold a conversation or even spread fake information at the push of a button.
However, Cade emphasized that technology has been promising people for more than half a century. It never quite gets there. Or not yet.
Shira: Explain how GPT-3 works.
Cade: It is based on neural networks. This is a technological approach that learns by analyzing large amounts of data. Siri recognizes spoken words and Facebook identifies faces in digital photos.
In this case, a neural network essentially searches the entire Internet to do one thing: predict the next word in a sequence of words. But it was also learned how people combine language. This means that you can generate tweets with very little prompt. Pen romance columns or even write simple computer programs.
That sounds crazy.
It’s incredibly impressive, but the system has significant flaws. If I were to try to generate a speech that sounds like President Trump, maybe five out of ten times the GPT-3 version doesn’t make sense would repeat sentences in ways people wouldn’t edit, and they would would have to be edited to make them logical.
If something only works half the time, it might not be that useful. People wrongly assume that technologies like this are more advanced than they really are because they choose the successful experiments and ignore the failures.
What is an example of something useful that GPT-3 could do and something that is potentially destructive?
An automated system that handles the repetitive, tedious parts of writing software programs could really make life easier for human programmers.
The most terrifying potential use is disinformation. Imagine a computer that can create an unlimited amount of false information or fake photos and that you cannot distinguish between real and fabricated ones.
Since the 1950s, scientists have repeatedly developed technologies that looked like we were close to mimicking human intelligence. Every time we were nowhere near.
The capabilities of this type of technology have improved significantly over the past 10 years due to neural networks, but this human mimicry is largely based on the identification and matching of patterns. People do much more than just match patterns. The question is when will the technology go beyond that.
Now, do these things become as capable or more capable than humans?
There is a massive argument about it even among artificial intelligence researchers. Some of them believe that if you give neural networks enough time and data or computerized simulations of the world, they will reach human intelligence. And others will find this just ridiculous, at least for the foreseeable future.
Will you and I still have writing jobs 10 years from now?
After using this stuff and talking to a lot of people about this technology, I’m not worried in the short term. But you can see the trend. This technology just keeps getting better and doesn’t need much further development to add to the human workforce or make certain work areas redundant over a long period of time.
I will say that language models like GPT-3 are like self-driving cars. They both improved dramatically, but they did last little improvement attaining human ability is very difficult.
TV apps are even worse
I packed on Tuesday via smartphone apps. However, streaming apps for television are even more terrible as they combine the worst aspects of the smartphone app stores and the cable television system.
What I mean is this: if you have cable or satellite television, every channel on your lineup is there by agreement between the channel owner and your television provider.
Walt Disney, the owner of ESPN, signs a complicated deal with Comcast, and Comcast pays Disney a ton of money, and then you can watch college football games. Sometimes these business deals reach an impasse and ESPN may go away for a while while Disney and Comcast fight for money. It’s annoying.
The streaming world could have rid us of this old school mess. Instead, it entrenched the chaos and took it one step higher.
Just like Apple and Google decide which apps will work on your phone, companies like Amazon and Roku decide which apps you can download to their streaming TV devices. And as with cable television companies – and unlike Apple and Google – Amazon, Roku, and their colleagues often enter into individual contracts with streaming services to split up the airtime for in-app commercials or for each Netflix subscriber who signs up, to pay a commission.
If the attorneys can’t make a deal, you may not be able too Find HBO max, the Peacock Video App or Apple TV + on your streaming device. It’s bad. That’s why you have to Consult a flowchart to watch TVand streaming apps probably cost more than they would have without this mess.
I do not know what to do. Maybe we should all plug in our smartphones or Computers to our televisions and bypass the streaming devices completely. But know that in your heart: the streaming entertainment system doesn’t make sense and it’s not your fault.
Before we go …
More unwanted attention for YouTube: Google’s own website banned right-wing broadcaster One America News Network for a week from posting new videos for breaking YouTube rules with false claims about a guaranteed cure for Covid-19. YouTube has specific guidelines to punish channels for spreading coronavirus-related misinformation, but not for making false claims of election fraud aggressively pursued by OAN, writes my colleague Dai Wakabayashi.
We can get rid of screens (a little): Set modest goals like a daily limit of 20 minutes for reading online news on the weekends, and create no-phone zones like the dining table. My colleague Brian X. Chen has these and others Tips on how to keep our minds from wallowing in our phones and other screens. (Counterpoint to Brian: when I’m not looking at screens I think of the world and get sad and give me all the screens.)
Use your precious screen time to stare at smooth brains: Wired is exploring the Memes that served as our attempts to laugh while screeching in a year full of disasters. There are the memes “Unfinkled Brain” and TikTok videos with dazed “nee noo nee nooSounds – anything to express that you don’t think about what’s going on around us.
Ping pong the cat fell four stories from their owner’s home to the roof of another building. Ping pong was rescued by London firefighters and she is fine! Although if you looked at it, you wouldn’t know Ping Pong’s expression in this photo. (Thanks to my colleague Erin McCann for sharing on Twitter.)
We want to hear from you. Tell us what you think of this newsletter and what else you would like us to explore. you can reach us at firstname.lastname@example.org.
If you don’t have this newsletter in your inbox yet, Please sign in here.