Comment Re:You sure about that? (Score 1) 124
Your problem is that "I lack the capacity to imagine such a system" is not the same as "such a system cannot exist".
Good day.
Comment Re:You sure about that? (Score 1) 124
unless you train your own model that you know has never seen the source, an endeavor that costs 10s of millions of dollars.
Look up "StarCoder" because it does not require anywhere close to what you claim it does to train from scratch.
Comment Ignorance or willful ignorance? (Score 2) 119
Don't buy the food, don't eat the food.
Are you ignorant or are you willfully ignorant?
Food has literally been engineered to be addictive. Don't believe the scientists? How about about looking at the food industry is responding by developing GLP-1 resistant food formulations. Given that obesity in the US has tripled in the last 50 years, if you don't suspect something odd is happening then you aren't looking.
Look at the facts, instead of trusting your "gut instinct" which you "know" is right because it's not.
Comment You sure about that? (Score 1) 124
You not only need to do a clean-room, you need to be able to prove it and that set-up does not allow it.
Actually, this setup would be more provable than having two people do it because you can literally record the entire process from start to finish.
Have two different computers with no hardware in common.
Computer 1 interprets the program and generates the documentation, saving it to a USB drive.
You unplug the USB drive and move it over to Computer 2.
Computer 2 reads the documentation and generates a new code base.
You can read the documentation and there was no other means of communication.
If you don't think a repeatable process is sufficient "proof" then you aren't being realistic and that's a problem with you, not the law.
Submission + - Python blood could hold the secret to healthy weight loss (colorado.edu)
fahrbot-bot writes: CU Boulder researchers are reporting that they have discovered an appetite-suppressing compound in python blood that helps the snakes consume enormous meals and go months without eating yet remain metabolically healthy. The findings were published in the journal Natural Metabolism on March 19, 2026.
Pythons can grow as big as a telephone pole, swallow an antelope whole, and go months or even years without eating—all while maintaining a healthy heart and plenty of muscle mass. In the hours after they eat, research has shown, their heart expands 25% and their metabolism speeds up 4,000-fold to help them digest their meal.
The team measured blood samples from ball pythons and Burmese pythons, fed once every 28 days, immediately after they ate a meal. In all, they found 208 metabolites that increased significantly after the pythons ate. One molecule, called para-tyramine-O-sulfate (pTOS) soared 1,000-fold.
Further studies, done with Baylor University researchers, showed that when they gave high doses of pTOS to obese or lean mice, it acted on the hypothalamus, the appetite center of the brain, prompting weight loss without causing gastrointestinal problems, muscle loss or declines in energy.
The study found that pTOS, which is produced by the snake’s gut bacteria, is not present in mice naturally. It is present in human urine at low levels and does increase somewhat after a meal. But because most research is done in mice or rats, pTOS has been overlooked.
Pythons can grow as big as a telephone pole, swallow an antelope whole, and go months or even years without eating—all while maintaining a healthy heart and plenty of muscle mass. In the hours after they eat, research has shown, their heart expands 25% and their metabolism speeds up 4,000-fold to help them digest their meal.
The team measured blood samples from ball pythons and Burmese pythons, fed once every 28 days, immediately after they ate a meal. In all, they found 208 metabolites that increased significantly after the pythons ate. One molecule, called para-tyramine-O-sulfate (pTOS) soared 1,000-fold.
Further studies, done with Baylor University researchers, showed that when they gave high doses of pTOS to obese or lean mice, it acted on the hypothalamus, the appetite center of the brain, prompting weight loss without causing gastrointestinal problems, muscle loss or declines in energy.
The study found that pTOS, which is produced by the snake’s gut bacteria, is not present in mice naturally. It is present in human urine at low levels and does increase somewhat after a meal. But because most research is done in mice or rats, pTOS has been overlooked.
Comment Re:The financial circle jerk's next phase (Score 2) 81
Bonus points for selling all the unsellable Cybertruck stock to SpaceX
I was unsure about this specific claim but it appears to be accurate: https://electrek.co/2025/10/13...
Comment Re:Not limited to open source. (Score 1) 124
While true, legally it makes no difference whether you steal the sources or the binary. It is still stolen.
Why would you steal it if you can simply license it?
And a clean-room implementation requires the code-writers to never have seen the original in any form. You cannot have an engineer analyze the original and then write a copy.
As the article explains, it's a clean room implementation because you use two different instances of an AI.
* AI 1 documents how the code works in a human readable descriptions. (i.e. does the reverse engineering)
* AI 2 constructs an entirely new codebase from the human readable descriptions in the documentation. (i.e. does the forward engineering)
Since AI 2 has never seen or analyzed the original code/binary and has only ever read the documentation about it, it is a clean room implementation.
Hence it is immediately plausible that having an AI train on the original or ingest it in a query and then writing a new version
The AI isn't training on the original (a very important point), it's generating written documentation on how it functions. The important part of this process is that it is creating a human level description and not simply an algebraic representation using words. Creating algebraic representation using words would simply result in generating a near identical source code which would be copyright infringement.
I do agree on the slop.
The current situation may or may not last but it's reasonable to assume that they are working on the problem of generating cleaner and more concise/less verbose code.
Submission + - Microsoft pulls faulty Windows 11 Update (techrepublic.com)
Ol Olsoc writes: Another Windows update has tripped over its own feet, and users are once again left staring at error screens instead of progress bars. Reports Tech Republic: https://www.techrepublic.com/a...
FTA: Microsoft has rolled back a recent non-security update after widespread installation failures. Designed to quietly improve performance and stability, it instead fails before it can even get off the ground. Unlike typical update issues that surface after installation, this one blocks users at the door, refusing to install or crashing midway through the process. For many, the result is a familiar frustration: a cryptic error message, stalled systems, and no clear path forward. Microsoft has since paused the rollout entirely while it investigates the issue, leaving users waiting for a fix instead of the improvements they were promised.
FTA: Microsoft has rolled back a recent non-security update after widespread installation failures. Designed to quietly improve performance and stability, it instead fails before it can even get off the ground. Unlike typical update issues that surface after installation, this one blocks users at the door, refusing to install or crashing midway through the process. For many, the result is a familiar frustration: a cryptic error message, stalled systems, and no clear path forward. Microsoft has since paused the rollout entirely while it investigates the issue, leaving users waiting for a fix instead of the improvements they were promised.
Comment Not limited to open source. (Score 3, Interesting) 124
Despite the open source spin, source code is not required to do this as source code can also be generated from binaries. It shouldn't be shocking by now to learn that you can fully automate breaking down executable into functional source code with the addition of AI to "make sense" of the generated code. As such, this means that even large sophisticated and complex programs are also targets.
The real question is, who wants to deal with a massive amount of AI slop code?
Comment Old news! (Score 5, Funny) 162
This technology has been around for decades. I know because my boss is definitely one of these brainless clones.
Comment It's already gone. (Score 4, Informative) 64
The github page that's being pointed to has already taken down the code. Unlike the fools that posted the WinAmp source code, they actually know how to wipe out the commits. However, I found that searching github with leaked Claude Code language:TypeScript was enough to find several mirrors of the code.
Submission + - We are nowhere near AGI (x.com)
schwit1 writes: Humans: 100%
Gemini 3.1 Pro: 0.37%
GPT 5.4: 0.26%
Opus 4.6: 0.25%
Grok-4.20: 0.00%
François Chollet just released ARC-AGI-3 — the hardest AI test ever created.
135 novel game environments. No instructions. No rules. No goals given.
Figure it out or fail.
Untrained humans solved every single one. Every frontier AI model scored below 1%.
Each environment was handcrafted by game designers. The AI gets dropped in and has to explore, discover what winning looks like, and adapt in real time.
The scoring punishes brute force. If a human needs 10 actions and the AI needs 100, the AI doesn't get 10%. It gets 1%. You can't throw more compute at this.
For context: ARC-AGI-1 is basically solved. Gemini scores 98% on it. ARC-AGI-2 went from 3% to 77% in under a year. Labs spent millions training on earlier versions.
ARC-AGI-3 resets the entire scoreboard to near zero.
Abstract and more here.
Gemini 3.1 Pro: 0.37%
GPT 5.4: 0.26%
Opus 4.6: 0.25%
Grok-4.20: 0.00%
François Chollet just released ARC-AGI-3 — the hardest AI test ever created.
135 novel game environments. No instructions. No rules. No goals given.
Figure it out or fail.
Untrained humans solved every single one. Every frontier AI model scored below 1%.
Each environment was handcrafted by game designers. The AI gets dropped in and has to explore, discover what winning looks like, and adapt in real time.
The scoring punishes brute force. If a human needs 10 actions and the AI needs 100, the AI doesn't get 10%. It gets 1%. You can't throw more compute at this.
For context: ARC-AGI-1 is basically solved. Gemini scores 98% on it. ARC-AGI-2 went from 3% to 77% in under a year. Labs spent millions training on earlier versions.
ARC-AGI-3 resets the entire scoreboard to near zero.
Abstract and more here.
Comment NOT friendly. (Score 4, Funny) 54
I was so proud of this robot that I went up and gave it a hug. Everyone around me started yelling at me to get away and then the robot tried to install a solar panel on me. Long story short, I ended up in the hospital with a restraining order against to stay away from the robot.
THIS ROBOT IS NOT FRIENDLY.
Comment Re:This is the QC we need! (Score 1) 18
You probably can't see how because of the same vision problem that prevented you from reading my post.
"simulate materials without ever having to construct them."
Comment This is the QC we need! (Score 4, Insightful) 18
Finally, we're seeing applications of QC that are actually useful. I look forward to this expanding so that we can greatly advance materials science by being able to simulate materials without ever having to construct them.
Slashdot Top Deals
I never cheated an honest man, only rascals. They wanted something for nothing. I gave them nothing for something. -- Joseph "Yellow Kid" Weil
