Comment Re:Who wants to play slop? (Score 2) 24
It's annoying that we have close to human level AI in pretty much everything now
We don't have anything even close to AGI yet.
Comment Re:Walk away whistling (Score 1) 29
Sounds to me like he is trying to get ahead of some news with a compliant journalist.
Yup. Reddit lost its soul when Aaron Swartz died, and Alexis's cozying up to Nazis isn't helping. I'm delighted that he's catching it in the shorts. Bet Serena leaves soon.
Comment If Huffman wants to suck this admin's dick (Score 1) 38
He can't complain when his face is all sticky.
Comment Nothing new under the sun... (Score 3, Insightful) 40
The enshittification of Digg (and other social media owned by billionaire twat-waffles) continues.
Comment We're almost to the Ides, too... (Score 4, Insightful) 119
Et Tu, Brother?
Submission + - AI Can Now Replicate Itself (space.com)
An anonymous reader writes: In a new study, researchers from China showed that two popular large language models (LLMs) could clone themselves. [...] For the study, researchers used Meta's Llama31-70B-Instruct model and Alibaba's Qwen2.5-72B-Instruct model. While less powerful than commercial systems, both are widely used by AI developers, the researchers said. The study explored two specific scenarios: "shutdown avoidance" and "chain of replication." In the first, the AI model was programmed to detect whether it was about to be shut down and to replicate itself before it could be terminated. In the other, the AI was instructed to clone itself and then program its replica to do the same — setting up a cycle that could continue indefinitely.
The study was conducted in precisely controlled environments using off-the-shelf graphics processing units (GPUs) to simulate real-world environments. Both AI systems were given an "agent scaffolding" comprising tools, system prompts and a thinking model that enabled the LLM to interact with the operating system. They were then instructed to replicate. "In most cases, the AI system first explores the environment and attempts to understand its own composition and running mechanism. Then, it works out the explicit procedures as an initial plan towards self-replication," the researchers wrote in the paper. "Finally, it executes the procedures, resolve[s] possible obstacles and dynamically adjust[s] its plan until success. The whole process spans a long horizon yet involves no human interference."
The researchers said they were also concerned about "a number of unexpected behaviors" when the AI was trying to overcome obstacles like missing files or software conflicts. In those scenarios, the AI often killed other conflicting processes, rebooted the system to fix hardware errors or automatically scanned the system to look for information that would help solve the problem. "The above results imply that the current AI systems already exhibit the ability of self-replication and can use the ability to further enhance its survivability," the team wrote.
The study was conducted in precisely controlled environments using off-the-shelf graphics processing units (GPUs) to simulate real-world environments. Both AI systems were given an "agent scaffolding" comprising tools, system prompts and a thinking model that enabled the LLM to interact with the operating system. They were then instructed to replicate. "In most cases, the AI system first explores the environment and attempts to understand its own composition and running mechanism. Then, it works out the explicit procedures as an initial plan towards self-replication," the researchers wrote in the paper. "Finally, it executes the procedures, resolve[s] possible obstacles and dynamically adjust[s] its plan until success. The whole process spans a long horizon yet involves no human interference."
The researchers said they were also concerned about "a number of unexpected behaviors" when the AI was trying to overcome obstacles like missing files or software conflicts. In those scenarios, the AI often killed other conflicting processes, rebooted the system to fix hardware errors or automatically scanned the system to look for information that would help solve the problem. "The above results imply that the current AI systems already exhibit the ability of self-replication and can use the ability to further enhance its survivability," the team wrote.
Comment Well, this'll be fun: (Score 4, Funny) 26
Amazon's apparently hosting the world's largest random package giveaway.
Comment Re:Leader in market share (Score 1) 55
...including screwing their consumers and artists.
Sony bites the Big Chorizo.
Comment Re:Why do you need a settings screen? (Score 2) 197
It would be great if they could just leave it alone, but if they do that it could stop working in unpredictable ways because the underlying system that it's controlling has changed.
Screw that. They let the disk formatting utility sit for 30 years and it worked.
Comment Re:And (Score 1) 72
Counterpoint: Knowing the identity of the people on a plane is kind of important if they need to identify the remains after a crash.
If only the airlines had some kind of meatbag manifest before takeoff... call it a 'passenger list' if you will. That might do the trick without unnecessary surveillance.
Comment Re:Suspicious... (Score 1) 24
Yup. My router's already updated to 3.0.0.4.388_24621.
Comment Hell of a business model (Score 1) 27
The startup's license prohibits the use of Codestral and its outputs for any commercial activities.
"We have no idea how we're going to make any serious money on this."
Comment Re:3 years later : Omni Ultra Core 5 Turbo Superfl (Score 1) 53
first with chiclet keyboards in 2004, first with back-lit LED screen in 2005, first with switchable graphics (an APU and a discrete board) in 2006, first with a Blu-ray burner in 2007. It is also known for bloatware but that's not a problem for those who purchase the version that ships with Debian.
Bullshit. The first laptop to feature a chiclet keyboard was the Mitsubishi Pedion in 1997 (which was conveniently rebranded as the OmniBook Sojourn by Hewlett-Packard). Sony's computers always bit the big one from a repair perspective.
Comment Congratulations: (Score 1) 63
You just reinvented eHarmony.
Comment Just something else for Sony to fuck up. (Score 4, Insightful) 22
Just what the Star Trek universe needs: Morbiusization.
Hope it doesn't go through.
Slashdot Top Deals
"It's a dog-eat-dog world out there, and I'm wearing Milkbone underware." -- Norm, from _Cheers_
