Such a click-baity name. Open challenge to anyone to use this framework and reverse engineer the internal APIs iPhone uses to talk to Siri or shortcuts or app Intents.
There's no obscurity anymore. Best you can do is keep as much as possible server side. Curious if anyone has experimented with LLM's RE'ing obfuscated code. I suspect they handle it just fine.
RE was arguably the first* task category to become completely solvable by LLMs on cybersecurity CTFs. This is exactly about obfuscated code, LLMs seem to have no problems with it now. The only probabilistic defense is to include tons of red herring so agent goes the wrong path and ends up in some tarpit like bruteforcing unrelated un-bruteforceable hashes.
* Cryptography category also went down fast but I think LLMs solved RE a bit faster. RE and crypto have in common that the task is completely local, but in classic RE tasks it is also usually ~obvious what you are supposed to do, you just need to do it - which agents do really well.
> Curious if anyone has experimented with LLM's RE'ing obfuscated code.
Sort of, albeit I haven't checked with SotA obfuscation, I've seen GPT 6.1 Sol find a URL hidden under a xorpad + 2 layers of encryption, and I've seen Astra RE an AES whitebox.
More importantly, remember these game decomps (e.g. Pokémon)? Well, with minimal guidance and IDA MCP these new models are able to do this _and_ properly name all function/params/locals in a single day.
today i read that for new game studios, the possibility of going stream only (like amazon luna) instead of downloadable is a probable future if they want to prevent piracy.
I ported a linux game binary from a Cortex-A7 to a Allwinner A133P (32bit to 64bit) in about two hours using Claude over the weekend. (I didn't have the source). SDL and the games encryption were the only things that made it more than a 15 minute activity.
If it has a binary value you can aim the Ai at, (aka does it work or not), the models are able to figure it out.
For laughs I'm probably going to have a Qwen model on my Mac min with the Pi harness do the same task in a week or two.
Hmm, prolly no reason this wouldn't work on older educational dos/windows games and porting them to linux. Damn, another weekend is going to be lost isn't it.
It seems both very lossy and very expensive (even in a future where agents are cheaper to run) to have an AI clone, say, a 3D video game. It also seems borderline impossible to have an AI usefully clone a service for which the served data is the valuable part, like TikTok or Facebook.
Such a click-baity name. Open challenge to anyone to use this framework and reverse engineer the internal APIs iPhone uses to talk to Siri or shortcuts or app Intents.
It will be interesting to see if anyone reverse engineers IDA Pro!
Eyy! Nice idea. I need to try to see how things work.
I dont know why Apple doesnt improve Siri. They have every resources on earth.
Steve had the vision …
FYI, Siri was effectively "fixed" as of OS 27. This is why Apple is finally able to launch their Siri-centric home products tomorrow. https://pogueman.substack.com/p/125-tests-of-the-new-ai-siri
It might reverse engineer them but that’s different from letting you use them
just curious, why is that hard to do?
focus and expertise. apple lacks both when it comes to AI. maybe it is strategic, maybe it is stupid hubris - we will soon figure out.
in any case, they're bound to get disrupted
Meanwhile they haven’t had iOS jailbroken since iOS 17 four years ago
Pegasus begs to differ!
There's no obscurity anymore. Best you can do is keep as much as possible server side. Curious if anyone has experimented with LLM's RE'ing obfuscated code. I suspect they handle it just fine.
RE was arguably the first* task category to become completely solvable by LLMs on cybersecurity CTFs. This is exactly about obfuscated code, LLMs seem to have no problems with it now. The only probabilistic defense is to include tons of red herring so agent goes the wrong path and ends up in some tarpit like bruteforcing unrelated un-bruteforceable hashes.
* Cryptography category also went down fast but I think LLMs solved RE a bit faster. RE and crypto have in common that the task is completely local, but in classic RE tasks it is also usually ~obvious what you are supposed to do, you just need to do it - which agents do really well.
Obscurity is kept by banning decompilation in EULAs. Can’t block black box reimplementation though (At least in EU), and that’s often good enough.
Obfuscation is no obstacle at all in my experience.
Are decompilation bans found to be widely enforceable or legal in most jurisdictions?
Genuine question; I do not know much about this area.
> Curious if anyone has experimented with LLM's RE'ing obfuscated code.
Sort of, albeit I haven't checked with SotA obfuscation, I've seen GPT 6.1 Sol find a URL hidden under a xorpad + 2 layers of encryption, and I've seen Astra RE an AES whitebox.
More importantly, remember these game decomps (e.g. Pokémon)? Well, with minimal guidance and IDA MCP these new models are able to do this _and_ properly name all function/params/locals in a single day.
today i read that for new game studios, the possibility of going stream only (like amazon luna) instead of downloadable is a probable future if they want to prevent piracy.
isn't this what google stadia was supposed to do and failed?
I ported a linux game binary from a Cortex-A7 to a Allwinner A133P (32bit to 64bit) in about two hours using Claude over the weekend. (I didn't have the source). SDL and the games encryption were the only things that made it more than a 15 minute activity.
If it has a binary value you can aim the Ai at, (aka does it work or not), the models are able to figure it out.
For laughs I'm probably going to have a Qwen model on my Mac min with the Pi harness do the same task in a week or two.
Hmm, prolly no reason this wouldn't work on older educational dos/windows games and porting them to linux. Damn, another weekend is going to be lost isn't it.
Brilliant. Time to take on the outrageous MSI windows software / broken Linux support.
this is going to lead to a trend of all software running on vendor computers instead of users its the only viable way
How would that protect anything? "Hey claude, clone that shit"
It seems both very lossy and very expensive (even in a future where agents are cheaper to run) to have an AI clone, say, a 3D video game. It also seems borderline impossible to have an AI usefully clone a service for which the served data is the valuable part, like TikTok or Facebook.
Facebook, TikTok, etc sell data, attention, infrastructure and established userbases, the software is just a medium.
[dupe] https://news.ycombinator.com/item?id=50028275
And now, we have to use a web browser for all of our professional software. fun times.