I used to be the Security Team Lead for Web Applications at one of the largest government data centers in the world but now I do mostly “source available” security mainly focusing on BSD. I’m on GitHub but I run a self-hosted Gogs (which gitea came from) git repo at Quadhelion Engineering Dev.
Well, on that server I tried to deny AI with Suricata, robots.txt, “NO AI” Licenses, Human Intelligence (HI) License links in the software, “NO AI” comments in posts everywhere on the Internet where my software was posted. Here is what I found today after having correlated all my logs of git clones or scrapes and traced them all back to IP/Company/Server.
Formerly having been loathe to even give my thinking pattern to a potential enemy I asked Perplexity AI questions specifically about BSD security, a very niche topic. Although there is a huge data pool here in general over many decades, my type of software is pretty unique, is buried as it does not come up on a GitHub search for BSD Security for two pages which is all most users will click, is very recent comparitively to the “dead pool” of old knowledge, and is fairly well recieved, yet not generally popular so GitHub Traffic Analysis is very useful.
The traceback and AI result analysis shows the following:
- GitHub cloning vs visitor activity in the Traffic tab DOES NOT MATCH any useful pattern for me the Engineer. Likelyhood of AI training rough estimate of my own repositories: 60% of clones are AI/Automata
- GitHub README.md is not licensable material and is a public document able to be trained on no matter what the software license, copyright, statements, or any technical measures used to dissuade/defeat it. a. I’m trying to see if tracking down whether any README.md no matter what the context is trainable; is a solvable engineering project considering my life constraints.
- Plagarisation of technical writing: Probable
- Theft of programming “snippets” or perhaps “single lines of code” and overall logic design pattern for that solution: Probable
- Supremely interesting choice of datasets used vs available, in summary use, but also checking for validation against other software and weighted upon reputation factors with “Coq” like proofing, GitHub “Stars”, Employer History?
- Even though I can see my own writing and formatting right out of my README.md the citation was to “Phoronix Forum” but that isn’t true. That’s like saying your post is “Tick Tock” said. I wrote that, a real flesh and blood human being took comparitvely massive amounts of time to do that. My birthname is there in the post 2 times [EDIT: post signature with my name no longer? Name not in “about” either hmm], in the repo, in the comments, all over the Internet.
[EDIT continued] Did it choose the Phoronix vector to that information because it was less attributable? It found my other repos in other ways. My Phoronix handle is the same name as GitHub username, where my handl is my name, easily inferable in any, as well as a biography link with my fullname in the about.[EDIT cont end]
You should test this out for yourself as I’m not going to take days or a week making a great presentation of a technical case. Check your own niche code, a specific code question of application, or make a mock repo with super niche stuff with lots of code in the README.md and then check it against AI every day until you see it.
P.S. I pulled up TabNine and tried to write Ruby so complicated and magically mashed, AI could offer me nothing, just as an AI obsucation/smartness test. You should try something similar to see what results you get.
Anything you put publicly on the internet in a well known format is likely to end up in a training set. It hasn’t been decided legally yet, but it’s very likely that training a model will fall under fair use. Commercial solutions go a step further and prevent exact 1:1 reproductions, which would likely settle any ambiguity. You can throw anti-AI licenses on it, but until it’s determined to be a violation of copyright, it is literally meaningless.
Also if you just hope to spam tab with any of the AI code generators and get good results, you’re not. That’s not how those work. Saying something like this just shows the world that you have no idea how to use the tool, not the quality of the tool itself. AI is a useful tool, it’s not a magic bullet.
I think that training models for fair use purposes, like education, not commercialization, will also fall under fair use. But even so, it’s very difficult to prove that someone has trained their model on your data without a license, so as long as it’s available, I’m sure that it’ll be used.
This “fair use” argument is excellent if used specifically in the context of “education, not commercialization”. Best one I’ve seen yet, actually.
The only problem is that
perplexity.aiisn’t marketing itself as educational, or as a commentary on the work, or as parody. They tout themselves as a search engine. They also have paid “pro” and “enterprise” plans. Do you think they’re specifically contextualizing their training data based on which user is asking the question? I absolutely do not.And this is why AI needs to be banned from use. People own the things they post / place them under various licenses, and AI coming along and taking what you did is a blatant violation of copyright, ownership, trust, and is just general theft.
I am absolutely angry with the concept of AI and have campaigned against its use and written at length, many times, to every company that believes it’s allowed to scour the internet for training data for its highly flawed, often incorrect, sometimes dangerous AI garbage. To hell with that and to hell with anyone who supports AI.
It hasn’t been decided in court yet, but it’s likely that AI training won’t be a considered copyright violation, especially if there is a measure in place to prevent exact 1:1 reproductions of the training material.
But even then, how is the questionable choices of some LLM trainers reason to ban all AI? There are some models that are trained exclusively on material that is explicitly licensed for this purpose. There’s nothing legally or morally dubious about training an LLM if the training material is all properly licensed, right?
Sounds like AI or an AI influencer post. The first paragaph is so far off-topic, might as well be talking about sailing. You completely mis-understood what I meant using TabNine. I wrote my own code and obfuscated my own code. Then tried to have AI complete another function using my code.
Nothing you said is relevant is any way, shape, or form.
[EDIT} https://www.tabnine.com/
My guy, your posts are particularly hard to follow, and you are very very quick to jump to the conclusion that you’re somehow being targeted and under attack. It’s no surprise that people aren’t responding to what you think is appropriate for them to respond to.
You’ve gone out of your way to provide extra info about irrelevant details: Why does the particular flavor of git you use matter at all to this conversation beyond the fact that you self host, why does it matter that you are on github as well when we are specifically discussing things you believe were sourced from readme.mds you have self hosted?
Meanwhile you don’t give many details or explanation about the core thing you are trying to discuss, seemingly expecting people to be able to just follow your ramblings.
Edit: After having re-read your OP, it’s less messy than I initially thought, but jesus christ man you need to work on arranging your points better. It shouldn’t take reading your main post, a few of your comments, and the main post again to get your point: “AI data scrapers appear to treat readme files as public data regardless of any anti-AI precautions or licensing you’ve tried to apply, and they appear to not only grab from github bit also from self-hosted git repositories.”
Seriously. OP might have a legitimate point but they’re making it with the energy of someone trying to convince me that vole people live in the antiposition of the time cube.
In fairness, a lot of the more exceptional engineers I’ve worked with couldn’t write their way out of a wet paper bag.
On top of that, even great technical writers are often bad at picking - or sticking with - an appropriate target audience.