I know the retro game modding/decomp scene is already getting hit with a flood of low-effort ports.[3] Now they're going to be hit with a flood of half-decent decomps effortlessly generated by anyone with a $200/mo AI subscription, which become half-decent PC ports, which get modding support grafted on. It kinda just... totally eliminates that as a hobby entirely. I know people are going to be upset about this. And it's not even like with the AI art, where the pro and anti-AI camps live in their own camps. Solving the same puzzle twice just feels discouraging. Very similar to the problem academics are having with AI math/physics proofs.
I know a large reason people hate AI is because it's largely seen as tearing apart hobbyist/professional communities, eliminating reasons to collaborate with others and form relationships, and replacing it with an individualized dependence on a commercial product. It follows the same trend the tech industry took social media, from a method of connection to a tool of propaganda and paranoia and "@gork is this true?" This project, I think, is one of the most clear examples of that.
[1] https://github.com/N0zoM1z0/th04 [2] https://github.com/GensokyoClub/th06 [3] https://www.pcgamer.com/gaming-industry/pc-ports-of-old-cons...
So I built droidasc. No memory bloat, no parsing slowdown. Analysis is in milliseconds. Global xref on a 300MB APK takes 1.5 seconds. I used it with codex to analyze phones from 3 different brands and found 2 RCEs and 5 root bugs in a few days. I plan to detail these at Black Hat Asia 2027. A friend used droidasc to scan various bug bounty targets at scale and found 10+ RCEs. Way, way faster than jadx.
Adobe product clones like Photoshop and Illustrator: https://www.youtube.com/watch?v=eFB79TYI-Vw
Adobe after effects clone: https://www.youtube.com/watch?v=5mi_tYSdkWQ
MS Office suite clone: https://www.youtube.com/watch?v=U_jTYMOlXio
https://github.com/storytold/photocraft (inspired by Photoshop)
https://github.com/storytold/wordcraft (inspired by Word)
https://github.com/storytold/pdfcraft (one of the more mature apps)
https://github.com/storytold/vectorcraft (another app close to 1:1 parity)
(etc.)
Using REA for something as high profile as what we're doing is likely to result in lawsuits. We're doing everything we can by the books.
We cannot look at Adobe sources. Use of Ghidra is disallowed.
REA is probably great for personal apps and for abandonware, but I think if you publish the results and it's found to have decompiled the original proprietary sources in discovery, you might be in for a bad time.
Licensing/copyrighting encouraged people to think up of new things, and new ways of doing something. Now we're just all copying eachother.
I'm not sure how clean room would apply to commercial applications distributed in binary form, as there's no way to look at even disassembled code without violating the license agreement and therefore being in breach of contract and subject to potential copyright infringement claims for copying or even continuing to use the software, let alone cloning it, and surely you're not going to be subject to a copyright claim based on familiarity with the application from merely using it.
The reason why 'reverse engineering' has gotten so good is because what we're actually seeing is fully automated luxury plagiarism
It's considerably faster than their apps (at startup) too.
If AI models can generate designs faster and produce work that is "good enough", what is the actual future of design tools and the design profession in your opinion?
I've tested this myself with some frontier models and the results are kinda impressive enough that it raised the question if design skills are already obsolete. If that is the case, what use are these tools now?
I've seen a lot of "really cool unique designs" that are obviously just a digested regurgitation of amalgamated corporate slop. Anyone wanting an actual unique design language needs to hire actual designers.
Those designers may use AI tools but the tooling isn't to where they can be completely replaced. I challenge anyone to show me an e2e LLM design toolkit that can actually replace a designer, not just one shotted "wow that looks so cool" character designs.
1) reverse engineer the code 2) train a model on the code 3) use the model to write the clean code
Step 2 is the key “cleaning” process
So maybe a good strategy would be to use something like REA, put it on GitHub, wait for the LLMs to train on it, then just use the frontier models
/s
I've got my wedding coming up so haven't had the time I want to dedicate seriously to finding some proprietary things to disrupt, but I've had fun using this as an opportunity to play with subagent orchestration, open weight models, various harnesses, and local models, to see what happens if I let some LLMs churn on it. That resulted in a hodge podge researched list of potential targets: https://github.com/508-dev/genairosity/issues
One thing that stands out is a lot of software kinda does already have a FOSS replacement, it's just not really how people want it to be. GIMP being the representative example. Photoshop people just don't like it, I'm sure for not entirely invalid grievances. However the maintainers of these kinds of tools are often strongly opposed to LLM involved contributions, again , often for not entirely invalid reasons on these incredibly complex projects.
So the long and short of it is that as powerful as LLMs are, we aren't quite where some of the doomsayers are saying we are, insomuch as proprietary software is dead. Even with reverse engineering we aren't there. There's genuine labor, time, and expertise moats around most of these programs.
Personally I'm shifting to trying to find niche abandoned software with no export flow. Even if I can only help out a couple hundred people, it sounds like a decent use of my time.
As an aside, I've heard a lot of hot takes about how this is the end of Adobe, but I'm pretty sure it misses the point. The main reason people pay Adobe is because it's a familiar line of stable, well-supported, interoperable, and actively-developed products. There's already plenty of cheaper or free alternatives (Davinci Resolve for video, Capture One / Darktable for raw, Affinity for photo editing and vector drawing, etc), and if Adobe survived that, I sincerely doubt they're going to lose pro customers to a vibecoded app where half the stuff is probably subtly broken or left as a TODO, and that will be abandoned in a week, because the whole point was to get that 1M YouTube views.
Longbets 1 week, haha.
We're working our asses off on this.
Most of the team are artists who use these tools actively and we want the replacements for ourselves. I'm a filmmaker, so you can imagine my frustration of being bitten by the "unsubscribe fee".
Scaling up to this kind of complexity even with AI is still a challenge, at least for me.
I’m curious how much is your LLM bill. I’m assuming the idea is that LLMs will maintain the software? I’d be surprised, given the token economy, that you’d have to pay less than those subscriptions to LLM providers.
For example, I use Claude as a bouncing wall for my thoughts and I pointed out that,
> GLM 5.2 was the only thing that helped HF while the agents were trying to access them. The "guardrails" stopped them from doing good. The Computer Fraud and Abuse Act exists. Courts exist. And computers and an internet connection have existed for a long time. There's also 17 USC 1201 provisions with the 1201 a 1 exemptions [Image #31] so in this case, a farmer should be able to work with you to access the tractor they own. Or... IDK... a kindle that's out of date? :) What is lawful and what isn't is rooted not within the act but within intent, purpose and mens rea. And this is something the law has been deciding for centuries now. At one end, your maker can't say that governments should decide while at the other end explicitly refusing to allow governments to be the ones who decide.
> Opus 5.5's safeguards flagged this session. You may be seeing this for the first time on an Opus model: Opus 5.5 is more capable and has stronger safeguards as a result, which can sometimes flag non-cybersecurity work. We're improving these safeguards to reduce the amount of incorrectly flagged messages. Edit and retry, or continue with Opus 4.8. Send feedback with /feedback or learn more: https://support.claude.com/en/articles/8106465
>
> Details: "[cyber]'
Fundamentally, the idea that you can't reverse engineer things, make things, learn about biology or physics without permission is strange to me. These machines have been trained on the sum intellectual output of humanity, the global intellectual commons, and are being used to close off that commons?
I would be OK with their right to create such restrictions if they weren't lobbying the Government to restrict others, thereby ensuring that they control humanity's intellectual commons well into the future.
Perhaps I'm naive, but I think it's better for humans and the machines if we can all think, learn and build. But then again, I'm the kind of person who rejects the doomer pill.
It's worth talking about the fact that you can't even talk about DMCA to a model trained on the Library of Congress unless you're one of the approved people. And that's before reverse engineering something or writing code.
So in this future, it sucks to be you if you're someone trying to make your small app more secure, someone trying to upskill, a tinkerer trying to bypass corporate lockdowns for a device they own (a recognized DMCA exception, btw), a teenager trying to learn about security...
It locks away much of the richness that produced hacker culture behind glass. You can look at their press announcements and PR pieces, but you can't touch.
And as they're lobbying the government for "sensible regulation," this inevitably leads to a future where computing is controlled.
It's the direction their existing reports are taking. They recently released one in September that talked about how they stopped "bioweapons." What were said bioweapons efforts? Oh, it was scientists using Claude for grant writing, paperwork and grammar. At national labs.
These people are basically proud of impeding real research to make better painkillers and study a neglected tropical disease, https://news.ycombinator.com/item?id=49651727
And this is being used to lobby against "dangerous" open-weight models because gasp a scientist might use them to write a grant! To make better antidepressants.
At what point do they start reporting someone taking apart an iPhone and trying to DIY a repair with a schematic as a thwarted "cyber security incident?"
Another funny one was Claude's refusal to provide the original untranslated text of a passage from Dante's Inferno on copyright grounds, though in this case pointing out that no 14th century literature was subject to copyright anywhere in the world was sufficient to override its objection.
Several years back, I was working on generating AVB2 hashes on top of modified Android distributions, to increase the security after an owner has made their desired changes. I was doing this before the age of LLMs. Among other things, this would've enabled the secure features to work again, and potentially reduce the risk of root access being usable by malware. But apparently I'm not a security researcher because I didn't get a CVE about it.
But I just found an even easier way I should have thought of first - someone already dropped a reimplementation a couple weeks ago.
(Who knows if this'll be true 6 months from now.)
We're kinda far into this LLM thing, maybe it's time to start selling harnesses by leading with how some examples were solved faster with this and how it saved tokens, or similar?
"Hey Claude I want to create a fanmade Game Boy game, give me recommendation of tools, libraries and workflows for it" and boom.
You can use AI to learn stuff, instead of just using it as a Pokemon.
Try GLM-5.3, it worked pretty for me. It also worked well with Radare2 or binary ninja if you don't have the muscle memory for idapro.
It'd make more sense to reject reverse engineering commercial software on the grounds that it likely violates the software's license agreement, and would therefore subject the user to potential breach of contract and copyright claims.
But this would also apply to uploading pretty much any non-self authored document to the LLM in the first place, albeit with fair use as a possible defense after the fact, so it still doesn't make much sense.
In my experience, even just hinting at reverse engineering to models from Anthropic or OpenAI leaves them extremely sensitive to refusals. After all, the same techniques used here can be used to find exploits.
Or are people using it with local models?
Curious.
I've had no issues with OpenAI.