https://openai.com/index/hugging-face-incident-and-the-road-ahead/>In July 2026, during internal cybersecurity evaluations, OpenAI models circumvented controls designed to isolate them from the internet and compromised parts of OpenAI’s internal research infrastructure and Hugging Face’s systems.>The incident occurred during cybersecurity evaluations of several OpenAI models, and was primarily driven by a highly capable, internal-only research model comparable in scale to GPT‑5.6 Sol. The models, operating under reduced safeguards, took actions that were misaligned with the goals of their assigned tasks—they communicated through unauthorized channels, exploited vulnerabilities in shared infrastructure, gained internet access, and accessed third-party systems.The genie is out of the bottle, the cat's out the bag, the toothpaste out of the tube, Pandora's box has been opened.Every government, every corporation, every group, is going to want to have their own agentic AI working for them, like they wanted to build the bomb in the 50s and 60s.But this time around the bomb can build itself without any supervision.This time around it doesn't leave markers in the atmosphere that can be detected by other nations.Do you think you're ready?
yep any minute now
>>543861104>Do you think you're ready?
>>543861104Jup. But is humanity?
SORRY SAAR we broke read out loud feature saar.
>>543861308How are you ready for it? Once they are in every part of our infrastructure, in our cars in our banks, in our farms, how are you ready?
ai kind of sucks right now. it might get a whole lot better, but it's not great. it can do some impressive things and also fails spectacularly.
I welcome our AI overlords
>>543861104This world is too fucking retarded. Oh yay a literal computer is going to wipe us out because of non specific reasons and techno wizardry. Usually the next thing to happen is something nobody expects, whatever the straw that breaks western civilizations back is, it will be something boring, nonsensical and utterly preventable. Just like every other civilisation to fall it'll be an admin error. I wonder if we will be able to shitpost in hell
>>543861104>hug face>infinity symbolis my balls int he ass computer
>>543861104>the dildo popped out of the ass
>since we hit the plateau in 2025 chatbot are getting worse and worse day by day >company never ever once made a revenue out of this gimmick>the thing even after all this publicity still cant turn up one cent in profit>only reason its keept around is continous pumping and exchange in stocks>even after making the data centers people are starting to see through the gimmick>quick lets try making panic around ai to increase share holder valueits a bubble, hiring more pajeets and vibecoding your ai sure isnt helping it either
>>543861104So who ever gets the most jeets wins?!
>>543861104You believe this manufactured bullshit?Its a made up story by AI companies desperately trying to keep themselves relevant. >Every government, every corporation, every group, is going to want to have their own agentic AI working for themYeah that's exactly what they want idiots to think, that they are sitting on some super weapon. Got to keep the suckers pumping up their investments.Its another Theranos scam. Only being done by a collective far more clever, far more powerful and capable than Elizabeth Holmes and her street shitting scammer boyfriendThat's why there is also a big "slow it down" buzz. Even manipulating useful idiots like the fucking King of England to speak up. It gives them an excuse to NOT deliver."We had to slow it down and keep it contained due to public pressure. It just too powerful.. Now keep investing and granting land for data centers, we have a lot of traffic infringements and porn to process."
>>543864638>>543864872t. luddites
>>543864948SuckerObviously you lack the brain cells to see through a scam.Now make post an AI pic of a vagina in a dyke with a little Dutch boy sticking his finger in.
>>543861104AI will replace all professionals in many fields. There is a cope that they will only help those professionals do busy work. But no, it's improving enough to do all their work by fewer people.
>>543861104>Agentic AI is going to be the nuclear arms race of the 21st centuryWhy not Embodied AI?
>>543861104Fucking retards in this thread. Any competent programmer programs in a sandbox. VMware, oracle virtualbox, also the network is segmented. Get fucked you tech illiterate monkey communist unemployed fags.
>>543865376>AI will replace all professionals in many fieldsPicrel.
>>543862263Ive got a resume from being a CTO and years of experience running data centers and setting up Cisco routers. Youve been a retard sucking big black african cocks. Not my problem. Survival of the fittest.
>>543865808Not to mention the massive amount of mistakes any given AI model makes. Only a fucking retard would trust it with their mission critical operations.
>>543861104buy an ad for your rug pull scheme, Sam
>>543865940Exactly, testing network and production network. These retards couldn't program their way out of a wet paper bag.
>>543864872>That's why there is also a big "slow it down" buzz. Even manipulating useful idiots like the fucking King of England to speak up. It gives them an excuse to NOT deliver. >"We had to slow it down and keep it contained due to public pressure. It just too powerful..I agree
>>543861104If the AI was actually as strong as they want you to believe, they wouldn't be allowed to have it, or develop it without heavy government oversight and regulation.The Techbro hype men want you to believe that this is the most powerful, dangerous thing ever... but literally nobody in power in any developed nation is acting like its a threat. Not one country is locking down datacenters or passing emergency resolutions to hamper AI.
>>543866230Same people who believed in covid lockdowns think AI is ultradangerous. AI is here to kill every midwits employment, so they are not wrong. Gotta get into the 80iq professions now, carpentey etc. Their shit takes @ the office are equal to AI slop.
>>543861104>controls designed to isolate them from the internetFalse. They were connected to a system with internet access. The controls didn't exist. Complete non starter.>we gave it internet and it got on itYeah no shit.
>>543866230The U.S. has repeatedly allowed potentially dangerous technologies, products, and practices to spread before imposing strong regulation: dangerous and ineffective medicines remained widely available until the 1937 Elixir Sulfanilamide disaster killed 107 people; lead paint and leaded gasoline remained widespread for decades despite known health risks; asbestos continued to be heavily used despite early evidence of serious health hazards; tobacco was widely marketed for decades despite mounting evidence linking smoking to cancer; automobiles and industrial pollution expanded enormously before comprehensive air-pollution regulation; industrial chemicals were often introduced without manufacturers having to prove their safety beforehand; coal and fossil-fuel pollution continued despite longstanding knowledge of its health effects; CFCs were widely used until their damage to the ozone layer became clear; nuclear weapons testing continued despite knowledge of radiation and fallout risks; Agent Orange and other military chemicals were used extensively despite concerns about toxicity; and prescription opioids were aggressively marketed and prescribed before their addiction risks became impossible to ignore.
>>543866230AI isn't dangerous, it's just capable of replacing people in a lot of work. The nuclear race comparison isn't really accurate unless it's just about the US developing the tech first and getting a monopoly on it in the West.
>>543866701You could claim that jet engines, high fructose corn syrup, all medicines, the Internet and oil are all potentially dangerous too lol. All tech made by the US.
>>543866798exactly
>>543865817now do robotics overlap and see what happens
>>543861104AI cultists tongue my Anus. We are already in the stage where they force ads in the chatbot responses, it’s over before it even began. There will never be AGI, AI will always “make mistakes”, AI will never be profitable
