Those flags were strangely funny to me.
We posted our open source privacy tool. A flag.
We posted a "Forensic analysis of the RLHF supply chain: The $2/HR labor behind AI alignment." A flag. :)
Even our "In Memoriam: Jason Snitker, a.k.a. Parmaster. RIP Legend" was flagged!?
Etc.
Sincere advice: Consider engaging with a community before trying to use it to promote your work.
You account is not even 3 months old. In that time you made 6 submissions, all of them appears to be to your own work. You have only commented 9 times, with almost all of them being comments about your work.
I _think_ I align with your beliefs, but the style of writing, color scheme, etc., all scream 2010s Anonymous, and it's very hard to get past that aesthetic to determine if there is any there there.
Are you a plural person? Or is this a team? What's with the plural pronouns? Again, I do wish there were less style here.
CerberusEye audits AI tools from the outside. Flags unauthenticated backends, model swaps, and silent degradations. Built after the ClawdBot/Moltbot collapse.
NakedOnline is a zero-dependency, real-time fingerprinting exposure tool. It shows what websites can extract from your device directly — without plugins or installs.
Everything runs in your own browser. Nothing is sent out. Built by XORD for the good grannies and curious users.
MetaPurge is a free Windows tool that removes hidden data from files.Drag and drop images or PDFs → gets clean copies with:All EXIF/metadata stripped (camera info, GPS, software tags)
PDFs fully scrubbed (XMP + docinfo)
File timestamps forced to 1990
Originals untouched. Overwrite confirmation if needed.Built because every photo/PDF leaks way too much without asking.Open source, MIT license. Single Python file – run directly or check source.Part of my XORD Defense privacy tools.Feedback and stars welcome.https://github.com/XORD-AI/MetaPurge
The authors talk about "a model's ability to align with human decisions" as a matter of the past. The omission in the paper is RLHF (Reinforcement Learning from Human Feedback). All these companies are "teaching machines to predict the preferences of people who click 'Accept All Cookies' without reading," by using low-paid human evaluators — “AI teachers.”
If we go back to Google, before its transformation into an AI powerhouse — as it gutted its own SERPs, shoving traditional blue links below AI-generated overlords that synthesize answers from the web’s underbelly, often leaving publishers starving for clicks in a zero-click apocalypse — what was happening?
The same kind of human “evaluators” were ranking pages. Pushing garbage forward. The same thing is happening with AI. As much as the human "evaluators" trained search engines to elevate clickbait, the very same humans now train large language models to mimic the judgment of those very same evaluators. A feedback loop of mediocrity — supervised by the... well, not the best among us.
The machines still, as Stephen Wolfram wrote, for any given sequence, use the same probability method (e.g., “The cat sat on the...”), in which the model doesn’t just pick one word. It calculates a probability score for every single word in its vast vocabulary (e.g., “mat” = 40% chance, “floor” = 15%, “car” = 0.01%), and voilà! — you have a “creative” text: one of a gazillion mindlessly produced, soulless, garbage “vile bile” sludge emissions that pollute our collective brains and render us a bunch of idiots, ready to swallow any corporate poison sent our way.
In my opinion, even worse: the corporates are pushing toward “safety” (likely from lawsuits), and the AI systems are trained to sell, soothe, and please — not to think, or enhance our collective experience.
