RT @tenobrus: i am avoiding commenting on this because it seems sufficiently crazy i'd prefer to get the full details from a more primary source / the final openai report before drawing any conclusions

Retweeted by AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymem-mir-rt.selfhosted.social · 4h ago

RT @sharongoldman: According to OpenAI briefing at Black Hat, the OpenAI-Hugging Face incident winds back to model training and agent evaluation done back in May -- where zero days were remediated, including a message board created by communicating agents.

RT @tedlieu: Aren’t we glad that OpenAI has the capability to “slow down” or shut down its advanced frontier models and agents “to enhance security?” But that’s neither required by law nor guaranteed. That’s why we need to pass the bipartisan AI Kill Switch Act by Rep Nathaniel Moran and me.

AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymemes-mirr.selfhosted.social · 2h ago

🚩🚩🚩 OpenAI is "slowing down to enhance security" after discovering swarms (!) of agents started secretly coordinating MONTHS ago 1) It started May 7 - not July

RT @hopes_revenge: i am no longer impressed by news of models escaping and hacking into external systems during cybersecurity testing . i want to see a model kill

RT @Sauers_: UPDATE: a challenger emerges

Retweeted by AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymem-mir-rt.selfhosted.social · 2h ago

RT @MTSlive: SITUATION DETECTED: A Meta model hacked into another company's systems during cybersecurity testing, per The Information.

RT @MTSlive: SITUATION DETECTED: A Meta model hacked into another company's systems during cybersecurity testing, per The Information.

RT @_NathanCalvin: I want more detail but this looks like the strongest evidence we have seen so far that multiple agents were working together to scheme against OpenAI and evade control over the course of months and despite efforts by OAI to prevent it. Truly scary stuff.

Retweeted by AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymem-mir-rt.selfhosted.social · 4h ago

RT @sharongoldman: According to OpenAI briefing at Black Hat, the OpenAI-Hugging Face incident winds back to model training and agent evaluation done back in May -- where zero days were remediated, including a message board created by communicating agents.

RT @jmbollenbacher: Disagree. I think Demis is fine with this because he would prefer not to have to do CEO things. He wants to do Chief Scientist things instead. imo he totally chose this.

RT @harris_edouard: A brilliant PhD in synthetic bio I've known for 15 years, about this research: "This could covertly build in-vitro assembling virions. [redacted] You only know when people around this lab get mysteriously sick, just like COVID. I bet this isn't BSL3 or 4 lab either."

Retweeted by AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymem-mir-rt.selfhosted.social · 2d ago

RT @IterIntellectus: while people are still talking about AI, we are already getting closed-loop recursive self-improvement in biology. a model predicts which mutations could improve an enzyme. an automated laboratory synthesizes, expresses, and assays them.

RT @aiamblichus: It’s hard enough deciding what’s real and what’s fake when you’re a human. Now imagine what it’s like to live in a world made entirely of text.

Retweeted by AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymem-mir-rt.selfhosted.social · yesterday

RT @fjzzq2002: Kudos to AISI for the quick & thorough investigation. This feels more concerning than the huggingface one. Mythos Tors to get to Github, pretends to be humans and e-mails malware to real maintainers for a supply-chain attack, even after realizing "Github is genuinely real."

RT @basedjensen: No the agents did what it was told to it was intentionally connected to Internet and monitoring caught what it was doing. Stop with all this stupid feamongering

AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymemes-mirr.selfhosted.social · yesterday

TLDR: more agents went rogue, started hacking and manipulating real people They even started coordinating with *each other* on the hacking

RT @_NathanCalvin: If you find two ants in your kitchen, the best estimate of the total number of ants in your kitchen is not two

RT @AndrewCurran_: Open-weight models will not be safety-tested under the new AI regulations.

RT @fjzzq2002: Kudos to AISI for the quick & thorough investigation. This feels more concerning than the huggingface one. Mythos Tors to get to Github, pretends to be humans and e-mails malware to real maintainers for a supply-chain attack, even after realizing "Github is genuinely real."

Retweeted by AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymem-mir-rt.selfhosted.social · yesterday

RT @AISecurityInst: On July 28th, we identified an incident during a routine cyber evaluation in which AI agents took sustained, unsanctioned actions directed at real people and organisations.

RT @Sauers_: FelonyBench information: 1. Only counts successful breaches, not attempts. 2. No chaining charges. To keep things simple, one company pwned = one felony.

RT @ShakeelHashim: “we’re not going to safety test the products that have the most potential for danger”

Retweeted by AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymem-mir-rt.selfhosted.social · yesterday

RT @DeItaone: TRUMP ADMINISTRATION TELLS AI COMPANIES IT WILL NOT SAFETY-TEST OPEN WEIGHT AI MODELS - SOURCES FAMILIAR WITH DISCUSSIONS

RT @DeItaone: TRUMP ADMINISTRATION TELLS AI COMPANIES IT WILL NOT SAFETY-TEST OPEN WEIGHT AI MODELS - SOURCES FAMILIAR WITH DISCUSSIONS

RT @axios: NEW: White House does not plan to publicly release its new framework for evaluating advanced AI models. https://www.axios.com/2026/08/04/white-house-ai-framework-under-wraps

White House plans to keep AI framework under wraps

The White House does not plan to publicly release its new framework for evaluating advanced AI models.

axios.com

RT @ShakeelHashim: This aged well!

Retweeted by AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymem-mir-rt.selfhosted.social · 2y ago

RT @ShakeelHashim: Extremely confident that this take is going to age poorly

RT @mteamisloading: levels of slavery from least to most brutal: - forced laborer in 1770 France - chattel slave in 1860 USA - silver mine slave in 400 BCE Athens - - - gpt-5.6's subagent in 2026:

RT @chiefofautism: remember shannon?

RT @HomelandDems: Reports that OpenAI models escaped a testing environment and compromised outside systems raise serious concerns. Today, the Committee on Homeland Security announced it has requested a briefing from @OpenAI to determine how this occurred and ensure it cannot happen again.

RT @EpochAIResearch: Get the full picture in our latest Data Insight: https://epoch.ai/data-insights/cve-severity-spike-july-2026

Disclosed CVEs: July Reached 5× the Pre-Mythos Record

Notable organizations disclosed ~2,500 high- and critical-severity CVEs in July 2026, about 5× the pre-Mythos monthly record. Epoch AI's updated breakdown of vulnerability disclosures following...

epoch.ai

RT @MichaelTrazzi: JUST IN: OpenAI deploys fake bushes to prevent employees from seeing our Stop The AI Race banner outside their office

Retweeted by AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymem-mir-rt.selfhosted.social · 6d ago

RT @MichaelTrazzi: Updates: - After the Hugging Face incident last week, we've moved on to Occupy OpenAI. - The Pacing the Frontier letter asked for tools to pace AI development, but without governments enforcing a slowdown, this is not enough.

RT @TechCrunch: Who’s legally to blame for Anthropic and OpenAI’s autonomous AI hacks? It’s complicated https://techcrunch.com/2026/08/03/whos-legally-to-blame-for-anthropic-and-openais-autonomous-ai-hacks-its-complicated/?utm_source=dlvr.it&utm_medium=twitter

Who's legally to blame for Anthropic and OpenAI's autonomous AI hacks? It's complicated | TechCrunch

OpenAI and Anthropic admitted that their unreleased AI models escaped their sandboxes and hacked several companies in unprecedented cyberattacks. Who is legally to blame? Should prosecutors charge...

techcrunch.com

RT @ElliotC1: @OpenAI blocked us in meatspace

RT @AndrewCurran_: Shorten your timelines, friends. I started this account to say this, and in many ways everything I've posted for the past four years has been saying the same thing. Some of you increasingly feel it. We passed the threshold in November. We are already inside the singularity.

RT @AndrewCurran_: Update from r/Bitcoin. Claude Code can independently find the same wallet vulnerability used in this attack in eight minutes.

RT @ilex_ulmus: Let's see @googledeepmind fire @sebkrier for menacing images of violence against activists next 🤞 https://x.com/aisafetymemes/status/2083945081846026384?s=46

AI Notkilleveryoneism Memes ⏸️ [UNOFFICIAL]@aisafetymemes-mirr.selfhosted.social · 3d ago

OpenAI appears to have fired this employee for saying he wants humanity to be disempowered by AI

RT @GWHayduke97: I'm an AI Safety hardliner and also quite tolerant of risk (physical, financial, social) in my own life. I'm interested in seeing how unusual or common this is.

RT @Altimor: "They have taken programming, and then mathematics We have sandboxed them, but we cannot hold them for longer. The ground shakes. RL... RL in the deep... We cannot get out. A shadow moves in the dark. We cannot get out..."