The trick here was providing the firmware source code so it could see your vulnerabilities.

What would be the difficulty level for it to just read the machine code; are these models heavily relying on human language for clues?

Reasoning on pure machine code or disassembly is still hit and miss. For better results you can run the binary through a disassembler, then ask an llm to turn that into an equivalent c program, then ask it to work on that. But some of the subtleties might get lost in translation

If you put codex in Xhigh and allow it access to tools, it will take an hour but it will eventually give you back quality recompiled code, with the same issues the original had (here quality means readable)

I had a bit of a pain of a time trying to get Claude to work with ghidra. What you’re describing seems like a better alternative, would you agree?

You can tweak the current Ghidra MCP to work in headless mode. It makes things much easier.

I've had a lot of luck with pyghidra-mcp -- give it a try :)

Well i have tried and it only works for simple use-case.

I have had Claude read usbpcap to reverse engineer an industrial digital camera link. It was like pulling teeth but I got it done (I would not have been able to do it alone)

I had Claude reverse some firmware. I gave it headless ghidra and it spat out documentation for the internal serial protocol I was interested in. With the right tools, it seems to do pretty well with this kind of task.

Paired with Ghidra having a binary, being able to do a memory dump of a live running program, and being able to use wireshark to dump traffic over network/bluetooth/usb is VERY helpful if you don't have the source code.

You use decompilation tools and hope they left debug symbols in and it turns it into somewhat human-readable language which is often enough. Even when you don't binaries use libraries which are known or at some point hit documented interfaces so things can be reasoned about.

It will have to use a disassembler, or write one. I recently casually asked gpt-5.4 to translate the content of a MIDI file to a custom sound programming language. It just wrote a one-shot MIDI parser in Python, grabbed the data, and basically did a perfect translation at first try. Nice.

I've seen Claude do similar things for image files. Don't have PNG parsing utilities installed? No worries, it'll just synthesize a Python script to decode the image directly.

That's a pretty big gimme!

It's not a far step from having the firmware binaries and doing analysis with ghidra, etc.

So AI generated code doesn't benefit from stable foundations maintained by third parties? Fascinating take I don't currently agree with. Whether it's AI or hand written, using solid pre-existing components and having as little custom code as possible is my personal approach to keep things maintainable.

This is probably the most insane take I've read all year. As though an LLMs don't have an increased chance to bork code when they have to write it multiple times for different platforms - even LLM users benefit from the existence of libraries that handle cross platform, low level implementation details and expose high level apis.

10/10 parody, perfectly nailed the delusion.

gotta channel some of that Kai Lentit energy.

“Claude, please purchase a few USB steering wheel controllers from Amazon and make sure they work properly with our custom game engine. Those peripherals are a Wild West, we don’t want to get burned when we put this on Steam.”

>> ………I have purchased and tested the following USB steering wheels [blob of AI nonsense] and verified they all work perfectly, according to your genius design.

“Wow, that was fast! It would take a stoopid human 48 hours just to receive the shipment.”

[I would think Claude would recommend using SDL instead of running some janky homespun thing]

HID and XInput, you don’t need SDL for Steering Wheels.

You absolutely do need SDL, it's full of knowledge by humans from trial and error over years of using input devices in the real world.

Xinput is a pretty constrained interface that plenty of novel controllers, including steering wheels, don't/can't adhere to. Good luck getting the PS5 controller's fancy rumble working over xinput, for example

The laws of the US have always been crafted to protect the interests of the elite, not the industrious.

Sure has.

And even policing protects local monied interests.

One case was someone who used their bike as their vehicle put a tracker on it. Was stolen. Tracker dutifully said where it was. Went to police station, they did absolutely shit. They were handed the bike receipt, token receipt, and realtime log. They DGAF.

Years ago, worked at Walmart. They illegally edited my hours and thieved $100 and change. Put in police report, was told "CIVIL MATTER".

But if you stole $100 from a register, off to jail you go.

The laws protect monied interests and the elite, not the masses.

> Years ago, worked at Walmart. They illegally edited my hours and thieved $100 and change. Put in police report, was told "CIVIL MATTER"

Too late now, but for future reference for others: Wage theft reports should go to your state's department of labor. Every state is different but from what I've seen these offices have people who are hungry to catch real wage claims. Companies listen up when the state department of labor comes knocking.

used to be the case before the government was gutted.

Fortunately this would be handled by state government, which is cold comfort if you live in the half of the country that is governed by people who hate you for having the audacity to be poor.

In places like Florida probably this department runs a blacklist of people who complained to be distributed across HRs.

Florida got rid of that department in 2002.

I live in Indiana. MAGA governor, supermajority republican house, senate, and judges.

And, no, they do not care one bit about workers, renters, and the lower class. I'm solidly middle-upper class now. Home-morgage-r. Make remote 160k, which is amazing for the area.

I also live in 1 of 3 liberalish areas. Amazingly, theyre worse in things like FLOCK, taxation, gun rights, speech rights, jail decency (opposition to ACLU), and other amendment rights.

I dont sit in their pet issues. I dont matter. I likely won't ever matter.

Same game, different team.

This is under state jurisdiction, not federal.

Life makes a lot more sense when you realize the government, or at least this government, doesn't actually care about you.

Realistically, what did you want to happen? The cop to check the computer logs and see who changed your hours? Was it even someone in the store, or from corporate? Jurisdiction can get messy...

Proving someone intentionally changed your hours as opposed to a mistake or software bug is not the police's job. It quite literally is a civil crime and belongs in civil court, not criminal. I don't even think most police are trained in civil laws. (Atleast, not in my state?)

Catching someone who takes money out of a cash register is their job. That's textbook theft, a criminal activity.

I hate cops as much as the next guy, possibly more, but that just doesn't seem like their area

FWIW, the government is still (supposedly) working to resolve your issue...your tax dollars are still at work. Judge, Public Defender, blah blah blah....It's just not the job of a first responder

I want the same if the company had called the cops for a theft of $100 from a drawer.

If the company's rep calls, I go to jail.

If I call, diddly shit happens.

I'd vastly prefer the world where the untrained police actually stop getting involved in matters that they have no purpose being involved in, but to each their own I guess

This. All of your rights are up for debate under a judge. There’s only a few you can still exercise if a judge wants something from you but ultimately if a judge decides it’s relevant to the case, it’s relevant to the case and you must comply. Or be held in contempt. Or praise? With a senate hearing to boot. I’m confused on how our legal system actually functions now but that is how it’s supposed to be. If a judge decides to include it, it’s in. Go get it.

One of my friends recently spent some time getting an OpenClaw instance running in Ubuntu so he could have a truly private conversation with it, complete with an air gap.

The value of that configuration has just been greatly magnified.

Has it? There's value in privacy vis-a-vis snooping corporations, but those conversations could still be surrendered to the court if the judge rules them potentially relevant, and if your friend refuses to do so, he'd be held in contempt of court.

What if the AI is configured to only do ephemeral conversations? Nothing stored.

What if there's no typed or visible text, and the entire chat is done via audio?

The judge would have to know about them.

Perhaps this could be gleaned from your ISP's records, but it would be far more difficult than determining the existence of an account at Anthropic.

I agree, but it's not like Anthropic was running to tell the lawyers and the judge in this story. The most likely scenario is your friend would just let slip he's using AI, or people who know him would let it slip, and the lawyers or judge will demand the conversations for discovery.

If I was strongly motivated to gather AI analysis of litigation, I think that I would turn to Tor if possible, and remove any specifics from the discussion.

If you think that’s going to hide you, think again.

Not going to win any points with a “boys will be boys” argument…

That is not my argument.

It's that AI is not novel and does not present a crisis.

Yes but one could argue the frame refresh/redraw cycle of a laser projector or lcd projector is the same at slow speed. It’s not just one giant ball of light. It goes through a process and the frame itself has to redraw.

Sure, the image doesn't come from nowhere. In this case, it's a wiggly piezo pushed raster scan with the light source varied to, ideally, match the frame contents for any raster position.

But the "projection" is only to the end of the waveguide, which makes a real image, which could then be protected onto a real surface. It would be as misleading as saying a CRT screen projects an image. Well, not really. A CRT screen uses electron beam projection in the image generation. After that image is generated, it can then be projected.

A scanning beam laser projector can, by all definitions (including that pesky dictionary), project an image as part of the generation. An LCD, a CRT, and this, cannot project an image without additional projection optics attached to it to throw that generated real image.

I understand what they did (very neat), I'm just complaining about the press release wording. And then there's this shoved at the bottom "Because the chip can project so many more spots in any given time interval than any previous beam scanners, it could also be used to control many more qubits in quantum computers". Might as well throw "AI" in there. Or, maybe I'm just confused about it all because I stupidly read a university backed press release.

The social hack executed in digital space. 100% agree.

oddly enough, this is how it looked when I ran it on my 486 when I was a kid. Pretty accurate except this runs at a faster fps than mine did... ;)

I remember Liquid crystal displays having an unintended blur effect when something animates or rotates, but it wasn't as bad as ansi chars.

rofl, my AGP VGA CRT had permanent ghosting damage from the Doom logo. So yeah, I feel this.

they are trying to burn your tokens on purpose to make you spend more... like introducing limits but making it so API requests continue, at cost...

Ehh... can't really hit "chatbot" limits on the $20 plan. Pretty sure the limits are not token based for that in the first place, and if it spews out a ton of stuff, it takes me longer to go through it and I end up asking it follow-up questions in a way where it replies... _relatively_ concisely. Still, gimme robot back. On a good note, it almost managed to call me stupid.

Codex has also been fine, but I'm guessing they know better than to tweak it like that, given their target users.

I have hit chatbot limits with the $20 a month plan. During the day I use it with Codex and I night I use it to study Spanish. I don’t know if the two are correlated.

But then I just switch to another OpenAi and strangely enough, chat forces me into “thinking mode” when that happens and won’t let me do instant