News for hackers. Stuff that matters.

mkatx

Re:Redesigning Source Control at Anthropic on Friday October 16, 2026 @11:32AM (#50119677)
That, and more commits means more data they can clean and train on, whether they're good or not. I'm sure they are commit-maxing and using AI as much as possible since they own that data and can use it however they want.
But I'd think at that point there'd be arguments for using that vps to instead facilitate a p2p connection... right?
Re:US Suspends Visa Program for Tech Firms Including Microsoft on Thursday October 08, 2026 @02:10PM (#50009605)
I wonder if wfh creates any loopholes here, and they just select areas where there are no real candidate pools.
Does anything else support Pascal gpu's though?
Re:Hole Punch: Sling your spaceship around gravitational fields on Saturday October 03, 2026 @03:01PM (#49946784)
Nice, a lot of fun!
Re:Revealing the details of how OpenAI agents hacked Hugging Face on Saturday September 26, 2026 @04:12PM (#49860113)
And the enumeration via fuzzing leads to reverse engineering.. so I argue it was a pretty smart path, both short term and long term.
> Snowden 2.0

"NSA ... AI Models"

Re:Fable 5 – Median thinking declined in August on Monday September 21, 2026 @06:33PM (#49794347)
Definitely noticed this before, but this parti6 time was very noticeable. I'm convinced it's to get the benchmarks in, then lower cost and prepare for the next release to look better relatively to users.
Re:English: A vs. An on Saturday September 19, 2026 @06:54PM (#49770798)
I always wondered if some uses of the word 'and' I notice on occasion might have similar explanations, or if they were the result of misspelling, mistyping, incorrect autocorrect, etc.. anyone know?

Examples from published articles: "Barely half and hour was on the clock" "the hospital, which is in and area of central Homs" "spiraled into and armed uprising and civil war" "During and interview with ABC in July"

Re:An Update on Wayback Machine Access on Tuesday September 15, 2026 @09:13PM (#49720998)
This is the way to go! Cut the cat and mouse, win win ish.
Re:Discovery of a new OpenAI agent message board on Saturday September 05, 2026 @07:29PM (#49581783)
You don't think they used their own AI, or Claude, to help build the sandbox, and that is where the failure lies?
Yep. Reading some of the earliest written stories really shows how little we have changed socially since then, and I imagine that goes back a lot further.
Re:DeepSeek-v4-flash-vision-exp on Friday August 21, 2026 @11:30AM (#49389634)
I would say, try without thinking on. I find reasoning on any rag type request seems to increase hallucinations, probably due to the thinking tokens taking attention away from the, in this case, vision tokens.

I'd recommend non-thinking for any non-prompt input, and leave the thinking where it has to actually reason.

Re:Cornell's Interactive Wall of Birds on Wednesday July 22, 2026 @11:01AM (#49007902)
Nice! I haven't tried Merlin, but I been meaning to set it up in the yard.

I've been using Seek on my phone, which is part of inatural. Similar thing, but pictures and most life, not just birds. That is totally Pokemon for non-pokemon peeps.. gotta bust out my phone every time I find something I haven't 'caught' yet.

Re:AI load breaks GitHub – why not other vendors? on Tuesday May 12, 2026 @08:21PM (#48116335)
Their rate limits are incredibly generous..

Edit: spelling

Just did a quick search.. dunno the vid quality, but the point being "'I Remove This Mysterious Tiny Chip Before Using The Phone!' Edward Snowden"

https://www.bitchute.com/video/QqQmHy6aqlLe/

I'm not sure how realistic this is, but conversations like this always remind me of this procedure.

Edit: maybe should have responded to the OP directly.

So instead of reverse engineering.. an llm/agent/whatever could simply produce custom apps for everyone, simply implementing the features an individual might want. A more viable path?
Re:Mullvad: Shutting down our search proxy Leta on Friday November 07, 2025 @09:50PM (#45853683)
Damn.. I just learned about this.
Re:We just hit $1M ARR in 4 years. With zero funding on Wednesday July 09, 2025 @12:52AM (#44506436)
Aw shit.

*drawing-board

Re:My 16-month theanine self-experiment on Sunday March 09, 2025 @05:54PM (#43314299)
Look into something like flexwatt tape. Maybe run it during the day when your not sleeping. In my mind, they probably use something similar inside heated blankets with some kind of limiter/thermostat.

Do your research though!! I'm not sure if flexwatt tape could ignite your mattress, but you would definitely want to know for sure yourself beforehand!

There are ways to control heat though; I'm familiar with using flexwatt tape with reptile enclosures, hooked up to a thermostat with a probe in the habitat for temp control. Maybe some math, like desired_temp = watts_supplied * feet_of_fw_tape, where the watts_supplied is the independent variable you adjust based on the required feet_of_fw_tape for your mattress to get the dependent variable desired_temp. Probably include a fail safe, but I'm no electrician, proceed at your own risk.

They do sell self contained fire extinguishers you can hang on your ceiling that deploy automatically at a certain temperature, ideally putting out fire sources they are hung above. Nice to have around for things like this.

Re:Inheriting is becoming nearly as important as working on Saturday March 01, 2025 @05:42PM (#43224617)
I do agree that changing policy is a separate, yet complimentary, step. But maybe the poor are disproportionately audited because the IRS doesn't have the funding to effectively target the rich and succeed? Maybe with better funding, they could successfully extract more tax dollars from the rich, incearsing the ROI of funding the agency?

I don't think further funding the agency is a completely nutty idea. Although if you truly believe the agency is broken, and that being broken isn't related to their funding, then I could understand your perspective. I lean more towards the idea that at least some, maybe even most, of the issues they have might be due to a lack of funding. That makes more sense to me than most of the criticisms I've heard of the IRS, which seem to mostly be on partisan idealism framed as otherwise.

Re:Fun facts about Google Scholar on Tuesday November 19, 2024 @04:14PM (#42188154)
You mean public, documented API's? Everything is/has an API.
Yeah. My favorite professor this semester constantly says "hey, if you rely to much on the robot, and can't do this yourself, you won't get a job." I know some people are just here for the paper, but that makes me feel better when I'm having a hard time finding a new role..
Re:All the data can be yours: reverse engineering APIs on Tuesday November 12, 2024 @04:01PM (#42119680)
I feel the same way about monetizing my api reverse engineering skills. I taught myself to program by reverse engineering api's, and it's what led me to pursue computer science. I'm pretty good at it, but I can't seem to find a role to flex my skills.
This is one big point I've subscribed to, I'd rather write the code and understand it that way, than read and try to understand code I did not write.

Also, I think it would be faster to write my own than try to fully understand others (LLM) code. I have developed my own ways of ensuring certain aspects of the code, like security, organization, and speed. Trying to knead out how those things are addressed in code I didn't write takes me longer.

Edit; spelling

Yeah, as a cs student, some professors allow use of LLM's because it is what will be a part of the job going forward. I get that, and I use them for learning, as opposed to internet searches, but I still manually write my code and fully understand it, cause I don't wanna miss out on those lessons. Otherwise I might not be able to verify an LLM's output.
Can confirm, 100% Texas right there. It is still worse for non-white men though, but even that won't save you here.
I see that, yes. But they will still create legislation against those means.
Wait, the ones who are getting a great deal on labor costs are the ones complaining about the source of that great deal?

I'm not so sure..

Re:Orion, Our First True Augmented Reality Glasses on Wednesday September 25, 2024 @07:22PM (#41652915)
So where do you sign up to be part of the "select external audiences access to Orion"?

Older »