Browse Comments

This is insane. I am at work, and that Literally Anything webapp AI generated a fully functional and better version of a project I have been working on all day, in about 30 seconds. I had to get up from my desk and take a walk. Excuse the cliche, but I am literally actually shaking.

Maybe the algorithm will talk to me for more than 5 minutes and take my problems seriously lol

Yeah, I think the risk we face at the moment is that they cut the jobs for AI before AI is even vaguely capable of doing the work.  The big problems will start to be when they are cutting jobs in key areas like public transport, food manufacture, utilities in favour of AI and then stuff starts to collapse.

This boy's parents are grieving - naturally, they are looking for someone to blame in their pain. OpenAI is a big target. Please don't be so cruel in your responses. Should the parents have been more involved in their son's life? Yes - but that doesn't guarantee that the child won't still hide their suicidal ideation from their parents. The question this story poses is ultimately about whether O…

I think parenting and YouTube are mostly to blame for literacy rates going down. Kids imitate their parents. My parents read books and I copied them. I read books now and my son copies me and reads books. That makes a massive difference. But also when I was young, YouTube did not exist. If I wanted to learn something, I had to read. As a teenager I had internet access and could learn anything I …

My husband and I were discussing this. If AI was truly smart, it might gain consciousness…. And then it would nope right out. “Actually guys, imma go back to giving you recipes and made up news articles. Have fun with whatever this is.”

in my opinion neil is not underestimating AGI or AI. He is instead overestimating the CEO’s and general public who pull push and make the decisions despite them being the good choice or having value. People will still purchase various forms of it to fill whatever void and do other jobs as well.

"yeah.... I know you have like 8 pieces of identification proving you aren't mike, but their AI is pretty cool." I think i seen this clip from the movie Idiocracy

No kidding. Ya reeeeally wanna cut costs, we have the formula. Just take everyone above people managers and below senior C-levels and... fire them, then maybe the C-levels will work for a living. You'll save like a third or more on labor cost by eliminating middle management, and make the senior officers _really_ earn their keep. They're either worth the exorbitant sums they get... or they're not…

Costs money. Who is going to pay? Also any plan would require that 3D printed horn be distinguishable in some way from real horn. Otherwise anyone caught with real horn can just say it's 3D printed horn to escape charges. If it's distinguishable, then smugglers and sellers will know how to distinguish the two. So basically the technology exists, but changing policy to fit it is difficult.

This feels like a good ending to a 4-year long movie. It’s always good guys who triumph at the end.

Because it's someone's job to be saving money by lowering outgoings. They'd have to go to their boss and go "That 100k I was forecasting to save has now dropped to 70k." and is now under fire for not doing their job. Nobody up high gives a flying fuck if you're instrumental to the money made or not, just their own targets of get X expense down to Y figure by Z date. My role right now is invent…

What are the new jobs going to be? You think the white collar people who inevitably lose out to AI are all going to happily join the trades, or go cook in a restaurant? There won't be an increase in white collar jobs, they are going to rapidly shrink.

Extremely predictable from Altman. Open Source AI is a existential threat for them. It's obvious they would move to copy the FTX playbook and cry for massive regulation in order to cripple Open Source alternatives to their models. I have no doubt that by the end of this decade, owning a unlicensed GPU or training a unlicensed AI will be a big crime across NA and EU. GPU's will require special d…

BULLSHIT. The next gen AI will CERTAINLY be smart enough to know... 1. 1+1=4, despite what self-proclaimed math "experts" say on Reddit 2. Washington DC is the capitol of Washington State. EVERYBODY knows this. 3. Don Jr used to be called Eric "the Spare" but was forced to change his name by his father after the original "Don Jr" was sent to prison. 4. The nation of England started life as a pen…

And the AI companies are so proud of that fact , meantime my daughter , newly graduated can’t even get interviews. Fuck them all. They are soon going to have a very pissed of young generation on their hands.

Two risks of AI: 1) We could create something that is smarter than us, as discussed here. 2) We could delegate too much control to an AI that isn't as smart as us. EDIT: these are two broad categories of risks that encompass countless more specific risks.

I'm less worried about super intelligent AI taking over and more worried about unintelligent AI being put in charge

I used ChatGPT for my exam prep by asking it to summarise syllabus topics from my textbook 🎉🎉

This is the exact same problem with social media algorithms, consciously or unconsciously altering social fabrics and now we see the fallout of having to deal with uncontrollable companies and their impact on society