Comment by smcl | Hacker News Reader

Comment by smcl | Hacker News Reader

> Back in 2014, Elon Musk referred to AI as summoning the demon. And it wasn’t hard to see that view. Soon, Go agents would beat top humans learning from self play. By the end of 2017, the same algorithm mastered Chess and Shogi. By 2020, it didn’t even need tons of calls to the simulator, and could play Atari too.

> AI looked scary. It looked like it was one FOOM away from self playing and becoming superhuman at the universe. And yet, here we are in 2023 and self driving cars still don’t work.

This is weird. I don't recall this at all. The mainstream press got a little kick out of chess and (to a lesser extent) Go AI turning over various humans at a few points over the years but it's only really burst into the mainstream recently. And where it did get any traction, in tech circles such as our own, response was enthusiastic but definitely more measured. Some were talking a bit about a kind of AI singularity way off into the future, but that was always a very distant and theoretical thing.

Westworld aired in 2016, Nolan's other TV show also featuring an AI world takeover plot is a little bit older still. If I had to count the amount of mainstream movie, video-game and television media that feature rogue AIs especially since the turn of the century I'd need a very long document. Hell go back a few more decades, WarGames, the Forbin Project. The literal term 'AI' wasn't as ubiquitous but threats from runaway automation are arguably mainstream for over half a century.

AI fears have been a cultural anxiety for a long time now and in many ways they're just a rehashed, secular version of Golem myths anyway.

The original article suggested an explosion of fears around the mid-2010s that our current AI tech was just one "FOOM" away from taking over the world. The fact that there have been a steady stream of sci-fi film and television programmes stretching back to the 80s does not support this.

> And yet, here we are in 2023 and self driving cars still don’t work.

Cruise and Waymo were today given licenses to operate across San Francisco 24/7.

Self driving cars do work and are here. They just need LiDAR.

It was the original article which made that claim, not me. However I think that two firms being given a license to operate in one specific city (which they've likely been testing in for a while) is not as enormous as you think it might be.

I think we can say that they're "here" when they're operating fully autonomously in multiple locales and landscapes. Maybe you're in SF so that Waymo/Cruise news means a lot to you, but I'll only say that once these firms are safely navigating narrow, winding streets in old European city centres that I am used to. I suspect if folks in, say, Accra or Tokyo or Montevideo or Islamabad or any number of other cities might feel the same way.

> really burst into the mainstream recently.

I think what made the LLMs so popular in the mainstream was anyone could go experience it themselves in a familiar, low barrier to entry way with a simple chat prompt.

IMHO LLMs are so popular because they sound very confident. That’s great marketing in itself. Unfortunately it’s only marketing.

Yeah you're right - I think "AI" as a concept for your average person was quite abstract and hard to wrap their head around, to get some kind of idea of the kind of things it can or cannot do. So having something tangible they can be a bit more hands-on with gives them a better idea of what it does.

Reply on news.ycombinator.com