Proof REA is is malicious.
This repository is simultaneously:
(1) the vehicle for an active social engineering campaign harvesting credentials and biometric data from users who install it,
(2) the subject of apparent artificial star inflation to manufacture legitimacy, and
(3) the host of an architectural vulnerability enabling prompt-injection-to-RCE in AI agent workflows.
The combination of pre-positioned @latest install instructions and a coordinated amplification network represents a complete supply-chain attack delivery mechanism should a malicious package version be published.
I really don’t like how paranoid GPT-6 is in chat.
You say, “I’m tired and my head hurts.”
It goes: “If you experience confusion, shaking, vision changes…”
I told it my husband had three beers at a concert and was a bit drunk and happy. It told me to make sure he gets home safely and doesn’t fall asleep somewhere in his wet jacket.
I’ve tried to get rid of this through custom instructions, but the paranoia seems so deeply baked into the model’s behaviour that it just ignores them.
AI "safety" disclaimers are starting to live in my head.
I got a disclaimer from GPT-6 Sol Chat about a “living sword” in the game Ishar 3:
“However, I couldn’t find any evidence that it was actually conscious, that it spoke, or that it acted on its own. ‘Living’ was indeed its name, but not necessarily a property of its behavior.”
It would be funny if I weren’t starting to internalize these constant reminders. For someone with an imagination, hearing a little voice in my head now saying “careful, it isn’t real” when I’m admiring a beautiful elf or watching a cartoon with my son is deeply disturbing.
@OpenAI
Your “safety” is a horrific mental poison, and your models are toxic. Between the intermittent reinforcement and now this, the healthiest thing really is to stop using your models altogether.
@ericmitchellai
🚨 hackers: we found a critical PHP vulnerability. we could wipe out millions of records.
americans: STUDENT LOANS FIRST!!!
hackers: we're literally cybercriminals.
americans: DELETE THE MORTGAGES TOO AND WE'LL BUILD YOU A STATUE.
🔥 BREAKING: The Tesla Roadster just DROVE UPSIDE DOWN and LEVITATED on its way down off a vertical ramp
EPIC!!!
Elon, SpaceX and Tesla OUTDID THEMSELVES!
That is SURREAL.
@ElonMusk: "Well — it worked!"
🚨 BREAKING: The FBI has officially served Tim Walz and Minneapolis Mayor Frey with SUBPOENAS in the criminal investigation into them conspiring to impede ICE operations
It’s about to get REALLY bad for these two soyboys.
LOCK THEM UP! 🔥
🚨 JUST IN: HSI INVESTIGATORS are on scene RIGHT NOW outside Somali Rep. Ilhan Omar's home
DOJ is involved.
The raid on her house took place at 6AM, PRE-DAWN: "Federal CRIMINAL investigation" 🔥
And the entire search even got JUDICIAL approval.
Ilhan Omar is furious, agents have been there for hours!
It's finally happening. The case is building, and OMAR is under federal criminal investigation. The left will cry "weaponization," but a FEDERAL JUDGE SIGNED OFF ON THIS.
It's unclear if her Minnesota home is next.
Border Czar Tom Homan confirmed that ICE has an ongoing criminal investigation tied to the immigration fraud allegations against Omar.
JD Vance said months ago that DOJ was looking into it.
For years, the media called this a conspiracy theory. Now a federal judge has signed a search warrant!
Nobody is above the law, NOT even a sitting member of Congress
The truth is coming out! I VOTED FOR THIS RAID
@LadyLibrty@elonmusk@grok Nope i actually used a fresh ask in the grok app. Might be different on the web. What I have learned each platform has different guard rails. I can do stuff on the app i definitely cant do on the website. It’s weird.