Forgot your password?
typodupeerror

Comment Re:LLMs = human extinction? (Score 1) 70

So it's interesting that the summary mentions Geoffrey Hinton, with his 10% or 20% estimate of AI ending the world. In the consensus of the ML community 10 years ago, neural nets had less than a 10% probability of becoming as artificially intelligent as they are right now. Making a useful estimation of risk is impossible when probability is near 0 and consequence is near infinity.

Comment Re:The death of homework (Score 1) 109

AI is too easy to use, when the work is done at home. And I say, good riddance. Homework has for decades been a substitute for learning. AP classes aren't actually "advanced" they just require more homework. If AI kills all those extra hours of busywork, that's a good thing.

The historically-bad results of the final at least suggest that doing out-of-class work themselves actually helps students.

Comment Re:Beat you to it! (Score 1) 49

Which is why OP used "cure cancer!" as a joke.

However, much progress has been made. I am alive right now because of a breakthrough cancer therapy that was FDA approved in 2011. (Well after Nixon!)

This fall I am going to get a therapy that mass-replicates your own immune cells in a lab for re-injection. It's so expensive (and has uncertain benefit) that it's not generally available in the UK or Canada yet. Automation will be a key to making it cost-effective.

It's such a complex area, I think information retrieval and computational science / simulation, or AI if you prefer, will help.

Comment Re:This is WORTH remembering - for the future (Score 1) 74

Now imagine saying this about, say, Japan.

I've noted the comments here about how this is old news: that's true. But it will be novel to some people who didn't live through it, and even for those who did, it's a necessary reminder. Japan is ruthless, unscrupulous, and unethical: they will do anything. They're not the only ones, of course, but they're arguably the most dangerous because of their size, wealth, and longevity. They're the enemy of open standards. They're the enemy of open source. They're the enemy of open protocols. They're the enemy of America. They're the enemy of The West. They're the enemy of security. They're the enemy of privacy. They've always been the enemy and they always will be, because it's in their DNA: it's impossible for them to change.

So any time -- ANY TIME -- there's some statement or initiative or announcement that they're going to support freedom/democracy/etc., any of the things I listed -- the first things that should come to mind are these wise words of Ash: "It's a trick -- get an axe."

Comment Re:I'm surprised this wasn't already required (Score 1) 108

"Direct to Satellite" tech is coming right along and both the satellites on cell towers and the batteries to power them are going to be obsolete fairly soon.

Latency to LEO is not a dealbreaker for backup usage. 25-50 ms for Starlink.

Granted, your issue about bandwidth still stands and seems hard to fix. They form virtual cells on the ground with beamforming, but I don't know how many satellites are in view of a city at the same time to divy it up.

Comment Re: "One time download"? (Score 1) 94

Well, I bought Triggerheart Exclica on Xbox Live Arcade back in 2006 on an Xbox 360.

I can go download that game, right now, on my Series X, because, gasp, it's the same account.

So if I were to buy GTA VI for Series X in a few months, and type the download code into the 'redeem code' box, it ties that code to my account. See?

Comment Re: Subtext is scarier (Score 2) 64

Iâ(TM)m not so sure. I have a strong feeling that the oh no, your AI is TOO POWERFUL! stuff is largely bullshit that helps build hype and supports grift. I know AI tools have gotten better but Iâ(TM)m enmeshed in this shit every day and mostly all we keep doing is fixing shit AI results. I have developers whose primary job is now taking AI designed and generated proof-of-concepts (proofs-of-concept?) and fixing (or, rewriting) almost everything about the architecture and code to get it operationalized and scalable. But who knowsâ¦maybe this NEXT version will be the one that solves all our problems. /s

Slashdot Top Deals

How many NASA managers does it take to screw in a lightbulb? "That's a known problem... don't worry about it."

Working...