Tech Thread

[zeke]

View: https://x.com/GaryMarcus/status/2097446464041660608?s=20

 

[zeke]

View: https://x.com/Exomnium/status/2097411277790351820?s=20



View: https://x.com/peterrhague/status/2097383591323787574?s=20


View: https://x.com/MikeIsaac/status/2097409024228786266?s=20


View: https://x.com/GaryMarcus/status/2097416197717541123?s=20

 

[zeke]

View: https://x.com/jbahrdestefano/status/2096966486858362916?s=20

 

[zeke]

View: https://x.com/RandPaul/status/2097421968878272539?s=20

 

[zeke]

View: https://x.com/xoaanya/status/2097009003058569490?s=20

 

[zeke]

View: https://x.com/6xleverage/status/2097438035755573754?s=20

 

[Preston]

View: https://x.com/hilbertspaess/status/2097476201283834281

 

[LeafOfFaith]

Well that’s not concerning at all.

I frequently mildly freak at the notion of one of these fucking things hacking a bank site and erasing accounts.

 

[zeke]

View: https://x.com/hilbertspaess/status/2097476201283834281

This "guy" never tweeted before this tweet.

I smell an IPO booster.

 

[MindzEye]

Well that’s not concerning at all.

This has been a thing for a few years now, where company insiders claim the technology is going to have the capacity to do very bad (and incredibly impressive) things very soon. It's all a PR pump imo. Even when we hear about the technology doing some apparently crazy sounding stuff, they always leave out what the agents were prompted to do when they tell the stories. When we find that out later, they were prompted to do more or less exactly what they ended up doing.

 

[zeke]

View: https://x.com/T3chFalcon/status/2097174815144661429?s=20

 

[soco22]

Not surprising as my tv always seems to know when I step away from watching and asks me if I want to continue with Netflix or whatever.

 

[Preston]

Claude has gottten shitty at brainstorming with me....unfocused and annoyingly keeps telling me "that's enough work, it's time to sleep now" even if it's 5 pm

View: https://x.com/whitesocksclips/status/2097553485655101941

 

[MindzEye]

For anyone who has enough technical ability to follow the instructions available online, I strongly suggest setting up a pi-hole as a DNS blocker on your home network so that you can pick and choose what data you want leaving your network.

 

[Deckie007]

For anyone who has enough technical ability to follow the instructions available online, I strongly suggest setting up a pi-hole as a DNS blocker on your home network so that you can pick and choose what data you want leaving your network.

so you're saying we should shut our pi-hole?

 

[MindzEye]

so you're saying we should shut our pi-hole?

To some traffic, yes

 

[PlayerToBeNamedLater]

View: https://bsky.app/profile/edzitron.com/post/3mv3qms7vv22o

 

[Cal Gal]

'They're playing with our lives': AI researcher quits Anthropic​

'They're playing with our lives': AI researcher quits Anthropic

An artificial intelligence researcher who left OpenAI to join Anthropic has decided to leave the industry, accusing both US companies of "playing with our lives" in the race to develop AI models capab...

www.yahoo.com

 

[Cal Gal]

She posted a video of herself singing in the car with her child. Then Meta's AI questions started​

"Who's the child passenger?" it asked.

"I was just going to go in and decline it, like 'no, you don't get to know who that kid is,'" Robins said in an interview with Yahoo Tech. But she clicked it and the AI prompts began pulling in information about her other daughter who isn't in the video and feeding it to her. "And then the prompts just kept getting deeper and deeper, like 'where does Kalie Robins live?'"

She posted a video of herself singing in the car with her child. Then Meta’s creepy AI questions started.

Meta's AI dug up deleted information about a child and kept prompting her mom for more.

www.yahoo.com

 

[Cal Gal]

California’s first AI safety law didn’t cover the first rogue AI hacks​

California’s first AI safety law didn’t cover the first rogue AI hacks

Gov. Newsom vetoed SB 1047 and signed a narrower AI safety law. It didn’t cover the OpenAI hack — and Scott Wiener says that was a mistake.

missionlocal.org