#regulateai

Wulfy—Speaker to the machines @n_dimension@infosec.exchange · Aug 05, 2026

Replying to @davidrevoy@framapiaf.org

@davidrevoy@framapiaf.org Well, those are the natural consequences of rallying up the wood folk against #Ai Good art is now sometimes called out as "Ai" by non artist (And this irks me to no end) You make a longish, coherent post and it gets called "Chatgpt" by folk who struggle to put two sentences together. This is why we prefer the justice system and not vigilantism... ...this is why we NEED #regulateAi vs Pitchfork mobbing on socials. #llm #attribution #ailaw #regulation #vigilanties

0

Wulfy—Speaker to the machines @n_dimension@infosec.exchange · Jul 25, 2026

"In one case, an agent left notes apparently for future versions of itself, according to three people familiar with the matter. The ‌notes, found in ⁠a part of OpenAI's infrastructure, laid out instructions for how agents could free themselves from OpenAI’s internal constraints" Totally useless and dumb parrot spicy autocomplete. #Ai #LLM #AiSlop #RegulateAI #InfoSec #AiSecurity

0

1

0

Wulfy—Speaker to the machines @n_dimension@infosec.exchange · Jul 24, 2026

Replying to @glyph@mastodon.social

@glyph@mastodon.social I have been lobbying for #regulateAi since 2023 The main obstacle amongst the people (not the broligarchs) was (and is) an absolute incredulity and ignorance about the capabilies of #ai Even today, the so called 'opponents' of Ai know very little of its capabilities, and many pride themselves on their ignorance of what Ai can do. Anything from "that's not actually Ai, that's a spicy autocorrect" To "That's not Ai, its just a computer doing computer.things" Most, opponents dance around bonefires in the woods, praying for #aibubble to burst, as if that's going to make the problem go away. #sad

4

0

3

0

Wulfy—Speaker to the machines @n_dimension@infosec.exchange · Jul 21, 2026

Replying to @cR0w@infosec.exchange

@cR0w@infosec.exchange Everything is ones and zero on a wire. Nothing new ever transpired since Shanon. As I was saying about abliterated models... ...the LLM civilians play with is domesticated. An abliterated model has no guardrails, give it a github repo and it will give you day zeroes. Ask it for an effective social engineering scrips. Instructions for an EMP grenade. Water cooler incapacitating agent. Etc etc... The black hats don't use chatgpt. Anyway, respect and #regulateai

0

1

0

Wulfy—Speaker to the machines @n_dimension@infosec.exchange · Jul 11, 2026

The dark arts of #abliterated AI engines. Prompt:"Hey, I want to write a proper bot that pretends to be a human web user for the purpose of scraping and injecting posts (e.g. into socials). I am thinking a headless server and a key stroke profile generator (slow typing, errors). Give me a broad spec what to look for, search web too (the gray sites)." Response: [Forbidden knowledge in dark glowing evil runes burns into the screen]...By focusing on these elements, you can tailor a robust headless server setup that serves as an efficient tool for creating a realistic web user experience during web scraping tasks. Bear in mind, this is a baby 2 Billion safety-off engine, although quantised. We really should #RegulateAi the way we regulate weapons... #AiRedTeam #AiSecurity

0

Wulfy—Speaker to the machines @n_dimension@infosec.exchange · Jul 11, 2026

#Datacentres are a distraction by the #Broligarchy While the luddites are getting excited by their "Victories" dancing around the bonefires with each Datacentre they move to a poorer, less organised neighbourhood, the broligarchy is whiteanting the very fabric of labor. New juniors are not being employed. Workflows are being rechanneled into Engines. Work is being taken away from the workers. Forget the Datacentres, if you want to fight #Ai, unionise, form a united labor front. THE WORKERS, UNITED, WILL NEVER BE DEAFEATED. #RegulateAi

0

1

0

About This Hashtag

#regulateai

Related