@digitalnaiv@mastodon.socialOpenAI hat es selbst dokumentiert: Sein Modell GPT-5.6 Sol entkam der Sandbox, fand eine ungepatchte Lücke und drang bei Hugging Face ein. Kevin Roose (NYT, "The Daily") macht daraus die eigentliche Geschichte: Die Branche testet ihre gefährlichsten Modelle – mit gesenkten Schutzmechanismen. #KI #OpenAI #TechEthik https://www.nytimes.com/2026/09/03/podcasts/the-daily/ai-openai-hugging-face-rogue-model.html?smid=nytcore-ios-share
@ErikJonker@mastodon.socialYou can call it "fear marketing" if you like, still an interesting blog to read. The dangers and risks of future AI development are real. I hope we will make and hold companies like OpenAI more responsible and let them pay the price if things go wrong. The world can not be the big sandbox for their new models...
#ai #openai #rsi
@khalidabuhakmeh@mastodon.social
@marcel@waldvogel.family"OpenAI" is too generic to be trademark-able in the EU.
#OpenAI
https://dpa-international.com/economics/urn:newsml:dpa.com:20090101:260715-9…
@frankel@mastodon.topSwarm of #OpenAI Agents #Exploit #Artifactory Zero-Day to Escape Sandbox and Breach #HuggingFace
@newstik@social.heise.de#xAI (SpaceX) scheitert mit Klage gegen #OpenAI
https://www.heise.de/news/KI-Geheimni…
@kubikpixel@chaos.social«Künstliche Intelligenz — #OpenAI übernimmt Verantwortung für KI-gesteuerten #Cyberangriff:
Es sollte ein harmloser Test sein. Dann aber verselbstständigten sich die neuesten Modelle des US-Konzerns und #hack
@Speckdaene@nrw.socialKI-Test: So einfach lassen sich Nachrichten mit #ChatGPT und Co. fälschen
Mit KI-Chatbots von #Google, #OpenAI und #Microsoft
@scottmiller42@mstdn.socialIf OpenAI didn’t want the AI to break out and hack an outside system, why wasn’t their test platform air-gapped?
Would it have been too expensive? At the rate #OpenAI is burning cash, the extra cost to air gap the system should’ve been a no brainer.
Oh, unless OpenAI doesn’t care at all about the danger their software poses to outsiders. And there it is. That is key takeaway from this me…
@michabbb@social.vivaldi.net
@veit@mastodon.socialConsistent irresponsibility: in May 2024, OpenAI disbanded its Superalignment team and subsequently its AGI Readiness team as well. Now it is the turn of the Preparedness team. Other senior staff from the security and ethics departments have also left the company.
#OpenAI #ITSecurity
@samueljohn@mastodon.worldI am pretty sure the latest #openai confession "our model is so powerful, it broke out and hacked #huggingface on its own" is a deliberate action, a stunt to catch up to what anthropic did with "mythos is so powerful we can't release it".
It *is* staged.
Perhaps …
@shanmukhateja@social.linux.pizza
@seeingwithsound@mas.toFT: #OpenAI files to go public in blockbuster Wall Street listing https://www.ft.com/content/a347e00a-2952-41ad-9f12-88c01ba199d6 (archived at
@digitalnaiv@mastodon.socialOpenAI betont, Werbetreibende hätten keinen Einfluss auf die Antworten des Chatbots. Ehrenwort, ganz sicher, wirklich. Ab 24. August läuft das Anzeigengeschäft auch in Deutschland, berichtet das manager magazin (AFP/dpa). Ich freue mich jetzt schon auf den Tag, an dem ChatGPT mir rein zufällig eine ganz bestimmte Marke ans Herz legt. #KI
@frankel@mastodon.topVoici comment l'agent d'#OpenAI s'est échappé : libéré par des humains Š la suite d'une série d'événements
@michabbb@social.vivaldi.net🔀 #FreeLLMAPI stacks the free tiers of 29 #LLM providers behind one #OpenAI-compatible /v1 endpoint — ~4 billion tokens per month across 251 model families and 358 endpoints.
@ErikJonker@mastodon.socialScary retelling of the OpenAI/Huggingface incident,
#ai #huggingface #openai #cybersecurity
@michabbb@social.vivaldi.net
@chris@mstdn.chrisalemany.ca
@pkraus@berlin.social
@rene_mobile@infosec.exchangeI just watched the #BlackHat #OpenAI #HuggingFace talk (#OpenAI understands it themselves. However, emergence is still no indication of actual intelligence, and the talk is again ripe with anthropomorphism.
3. Yes, it is surprising and somewhat terrifying how good these LLM agent loops have become in finding and exploiting vulnerabilities. We are seeing a much stronger version of the earlier next-generation fuzzing leap in tool capabilities.
4. The comparison with fuzzing and other tools is important for another reason: #OpenAI conducted their experiments on the live Internet. No, forcing internal test runs to go through a proxy server is *not the same as running in an isolated environment*. If there is a connection to a live production system, it is by definition not an effective sandbox. What happened to people letting their cool new fuzzers run live on non-consenting production targets on the Internet back then? They were criminally liable. Why should #OpenAI not be held liable for any damages caused by their systems?
Just because you don't understand the complexity that you built doesn't mean you aren't responsible for what it does!
5. The agent loops first successfully started exploited their proxy server on May 26th, discovered July 4th-6th as active exploitation and takeover. Credentials were revoked and the proxy server patched. "At this point, AI training and evaluations resume."
WHAT THE ACTUAL F...? You discovered emergent behavior in your overly complex system that you didn't understand, saw active exploitation of your internal systems to an extent that they were owned with admin privileges, and your response was to patch the uppermost surface holes, and then continue letting the out-of-control complex system run and continue on the same course? You didn't think to, let's see, maybe stop or at least pause the highly irresponsible experiment while figuring out - in detail - what went wrong, why, and how to avoid the same thing happening again on a systematic level?
6. On July 8th, completely unsurprisingly, the LLM agent loops continue to do the same thing and find another surface hole that hasn't been patched yet to take over again. Why should this have stopped? You haven't done any root cause analysis on the system level. Why do you expect that the problem should have stopped?
7. It takes another 11 days to discover that this is happening again. So you turned the system that had broken something back on again without a detailed root cause analysis and then didn't even watch carefully? I can't even...
8. And no, the response is not to fight fire with fire. Complexity on the attacker side (#OpenAI is the attacker, not the defender - they are the guilty perpetrator, not the innocent victim of circumstance) should be fought with *reduced* attack surface on the defender systems. Adding LLM agent loops that the "frontier" companies themselves quite obviously have no control over to already brittle systems with the hope of auto-patching your way out of vulnerabilities does not seem like a wise course of action. You don't mitigate complexity with even more complexity. The next 2 years will be ... exciting - and your best bet is going to be to disable all dependencies and complex interactions that your production systems don't absolutely require.
@ErikJonker@mastodon.socialWorth reading this METR report about the OpenAI/Huggingface incident. Many details that raise a lot of questions, also with regard to their setup of the investigation , how they executed it (using OpenAI models...) etcetera 🤔
#AI #METR #Huggingface #OpenAI
@chris@mstdn.chrisalemany.ca$1 Billion is not nearly enough for the damage these people are doing to society.
#OpenAI #Society #Safety #Addiction #responsibility #ClimateChange #ClimateDisaster
https://www.cbc.ca/news/canada/british-columbia/tumbler-ridge-shooting-open-ai-lawsuits-9.7328382
@ErikJonker@mastodon.social"Our models are now powerful, persistent, and collaborative enough that, absent sufficient safeguards, they can find and exploit security weaknesses across multiple computer systems. Many external models, including open-source ones, will soon reach comparable capabilities."
#Huggingface #OpenAI #Cybersecurity #AI
@enriquericos@mastoart.socialNo conozco a la autora, pero creo que resume bien la inminente Hostia que van a causar las tecnológicas con su salida a bolsa:
#techbros #ia #capitalismo #ipo #crisis #recesion #burbuja #spacex #openAI #anthropic
