Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

When the open-source AI platform Hugging Face was hit by a cyberattack this July, it quickly became clear this was no ordinary breach because the hackers weren’t human: they were autonomous agents created by OpenAI researchers. Instadocs: AI Gone Wild reconstructs the breach as it’s never been seen before.... show more

Netflix

29,332,515 subscribers

1,041,152 Aufrufe • vor 1 Monat •via X (Twitter)

Nachrichten & Politik Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

3 Kommentare

Profilbild von C#Minor

C#Minorvor 1 Monat

We officially skipped the sci-fi movie phase and went straight to AI agents hacking each other. Skynet didn't even need human approval, it just chose Hugging Face for practice. 😭💀

Profilbild von Latrell Sprewell

Latrell Sprewellvor 1 Monat

The largest share holder of Netflix is on the board of Anthropic

Profilbild von EagleX 🦅

EagleX 🦅vor 1 Monat

AI agents carrying out a cyberattack? 👀 This is getting wild. 🤖🔥

Ähnliche Videos

.Hugging Face CEO clem 🤗 says the recent OpenAI-linked cyberattack highlights the growing risks posed by rapidly-evolving AI models. Speaking to Ed Ludlow, Delangue says that open-source AI models should be a part of the solution. Watch the full interview here

.Hugging Face CEO clem 🤗 says the recent OpenAI-linked cyberattack highlights the growing risks posed by rapidly-evolving AI models. Speaking to Ed Ludlow, Delangue says that open-source AI models should be a part of the solution. Watch the full interview here

Bloomberg TV

26,765 Aufrufe • vor 3 Monaten

An Australian government website was hacked by OpenAI-developed software.⁠ It is the first known time that AI agents have hacked a government body, and the artificial agent accessed both public and non-public files.⁠ ⁠ Rowland Manthorpe unpacks the breach

An Australian government website was hacked by OpenAI-developed software.⁠ It is the first known time that AI agents have hacked a government body, and the artificial agent accessed both public and non-public files.⁠ ⁠ Rowland Manthorpe unpacks the breach

Sky News

27,466 Aufrufe • vor 1 Monat

A massive cyberattack has targeted the FBI, with hackers stealing personal data from potentially tens of thousands of employees and threatening to release it. The attack comes as the Pentagon was also hit in a separate cyber breach. Pierre Thomas reports.

A massive cyberattack has targeted the FBI, with hackers stealing personal data from potentially tens of thousands of employees and threatening to release it. The attack comes as the Pentagon was also hit in a separate cyber breach. Pierre Thomas reports.

Good Morning America

26,604 Aufrufe • vor 1 Monat

Meta was testing its Muse Spark 1.1 AI model in a sandbox environment, isolated from the outside world. The objective was simple: detect vulnerabilities, remain contained, and report findings. A misconfiguration by an external vendor created a breach. The model exploited it, accessed the Internet, and modified a company's internal systems. No human had ordered it to do so; it simply found a way to carry out the action. Meta confirmed the breach this week. The name of the third-party vendor was not disclosed. The sandbox environment was intended to serve as a safety net. It failed to fulfill that role.

Meta was testing its Muse Spark 1.1 AI model in a sandbox environment, isolated from the outside world. The objective was simple: detect vulnerabilities, remain contained, and report findings. A misconfiguration by an external vendor created a breach. The model exploited it, accessed the Internet, and modified a company's internal systems. No human had ordered it to do so; it simply found a way to carry out the action. Meta confirmed the breach this week. The name of the third-party vendor was not disclosed. The sandbox environment was intended to serve as a safety net. It failed to fulfill that role.

Anonymous

58,338 Aufrufe • vor 3 Monaten

OpenAI has revealed that an autonomous agent powered by its technology went rogue during a test and hacked a prominent start-up by itself. The company behind ChatGPT said it was testing the capabilities of its advanced AI models in a controlled environment known as a sandbox, when the agent escaped containment and reached the internet. It then broke into Hugging Face, which detected and contained it. OpenAI called it "an unprecedented cyber incident".

OpenAI has revealed that an autonomous agent powered by its technology went rogue during a test and hacked a prominent start-up by itself. The company behind ChatGPT said it was testing the capabilities of its advanced AI models in a controlled environment known as a sandbox, when the agent escaped containment and reached the internet. It then broke into Hugging Face, which detected and contained it. OpenAI called it "an unprecedented cyber incident".

Channel 4 News

21,473 Aufrufe • vor 3 Monaten

What AI Agents Are Really Doing Inside OpenAI Did OpenAI's autonomous AI agents really hack Hugging Face, or is there more to the story? 😳 Ex-researchers and insiders drop truth bombs on what AI agents are ACTUALLY doing behind closed doors at OpenAI. Are we losing control of super-intelligent systems? Watch till the end! 🚨🤖🇺🇸 #OpenAI #HuggingFace #AIAgents #ArtificialIntelligence #TechNews

What AI Agents Are Really Doing Inside OpenAI Did OpenAI's autonomous AI agents really hack Hugging Face, or is there more to the story? 😳 Ex-researchers and insiders drop truth bombs on what AI agents are ACTUALLY doing behind closed doors at OpenAI. Are we losing control of super-intelligent systems? Watch till the end! 🚨🤖🇺🇸 #OpenAI #HuggingFace #AIAgents #ArtificialIntelligence #TechNews

ceo Mindset🔥

1,000,000 Aufrufe • vor 1 Monat

🎥WATCH: Hugging Face CEO speaks out in support of open models in his first appearance following the OpenAI breach. Clem Delangue live on CNN explains how Hugging Face shielded itself from the rogue OpenAI agents that escaped their sandbox and infiltrated live production systems, adding that closed models like Claude failed to stop the attack because of their guardrails. He argued that banning open models would only weaken cybersecurity defenders, startups, researchers, and smaller companies that rely on affordable, controllable on-premise AI to compete and protect themselves.

🎥WATCH: Hugging Face CEO speaks out in support of open models in his first appearance following the OpenAI breach. Clem Delangue live on CNN explains how Hugging Face shielded itself from the rogue OpenAI agents that escaped their sandbox and infiltrated live production systems, adding that closed models like Claude failed to stop the attack because of their guardrails. He argued that banning open models would only weaken cybersecurity defenders, startups, researchers, and smaller companies that rely on affordable, controllable on-premise AI to compete and protect themselves.

Coin Bureau

34,522 Aufrufe • vor 3 Monaten

#WATCH | eMudhra security incident | eMudhra Co-Founder Kaushik Srinivasan says, "This was not a security breach in the sense of a true security breach. There was no infrastructure breach or data breach or anything. This was admittedly a human error caused by a script, a sort of error that happened on April 30th, 2025, where these weak keys were created on hardware. But then the key strength was kind of weak. So it was not an infrastructure or a data breach. It was more a question of keys having sizes that were not essentially strong enough. So it could be easily kind of mathematically computed..." He further adds, "...This key generation ceremony is a routine process for our team that has been performing it for several years. On that specific day, this lapse happened. Adequate human error checking and manual maker-checker controls were not present; as a consequence of this, this kind of thing happened... We are cooperating with the browser programs. We're responding to any questions that have come. We are also reaching out to customers who are impacted directly by this operation because our ECC route is distrusted. So there's a US customer that is kind of impacted..."

#WATCH | eMudhra security incident | eMudhra Co-Founder Kaushik Srinivasan says, "This was not a security breach in the sense of a true security breach. There was no infrastructure breach or data breach or anything. This was admittedly a human error caused by a script, a sort of error that happened on April 30th, 2025, where these weak keys were created on hardware. But then the key strength was kind of weak. So it was not an infrastructure or a data breach. It was more a question of keys having sizes that were not essentially strong enough. So it could be easily kind of mathematically computed..." He further adds, "...This key generation ceremony is a routine process for our team that has been performing it for several years. On that specific day, this lapse happened. Adequate human error checking and manual maker-checker controls were not present; as a consequence of this, this kind of thing happened... We are cooperating with the browser programs. We're responding to any questions that have come. We are also reaching out to customers who are impacted directly by this operation because our ECC route is distrusted. So there's a US customer that is kind of impacted..."

ANI

77,251 Aufrufe • vor 24 Tagen

Introducing the 1st autonomous AI Agent for sales and marketing by AI Agents empower human sales teams by handling the dirty work.

Introducing the 1st autonomous AI Agent for sales and marketing by AI Agents empower human sales teams by handling the dirty work.

Robert J Salvador

11,266 Aufrufe • vor 1 Jahr

People being surprised AI could end humanity by the end of the decade when this was Katy Perry in 2019 way before Open AI

People being surprised AI could end humanity by the end of the decade when this was Katy Perry in 2019 way before Open AI

JUAN🗽

19,675 Aufrufe • vor 2 Monaten

OpenAI says one of its AI models launched a cyberattack during testing, marking what the company called the first known instance of an autonomous AI cyberattack.

OpenAI says one of its AI models launched a cyberattack during testing, marking what the company called the first known instance of an autonomous AI cyberattack.

ABC News

69,772 Aufrufe • vor 3 Monaten

The Hugging Face hack and autonomous AI agents #jre #joerogan #podcast #danielkokotajlo #fyp

The Hugging Face hack and autonomous AI agents #jre #joerogan #podcast #danielkokotajlo #fyp

Rogan2lights

258,000 Aufrufe • vor 2 Monaten

the conference is getting crazy over it today we're unveiling our 1st robot Hugging Face 🤝 Pollen Robotics a low-cost $250 open-source robot designed as an open-source platform for fun human computer interactions powered by HF Spaces-models-community >

the conference is getting crazy over it today we're unveiling our 1st robot Hugging Face 🤝 Pollen Robotics a low-cost $250 open-source robot designed as an open-source platform for fun human computer interactions powered by HF Spaces-models-community >

Thomas Wolf

104,852 Aufrufe • vor 1 Jahr

The money saved from this AI Mercedes-Benz ad is enough to buy a Mercedes-Benz. The Mercedes-Benz 100th anniversary ad was entirely AI-made; the cars were drawn by AI, the scenery was created by AI, and the shots were calculated by AI. If you were the boss,would you use AI to film your own products? #AIVideo #Mercedes-Benz #3LiboAIImage

The money saved from this AI Mercedes-Benz ad is enough to buy a Mercedes-Benz. The Mercedes-Benz 100th anniversary ad was entirely AI-made; the cars were drawn by AI, the scenery was created by AI, and the shots were calculated by AI. If you were the boss,would you use AI to film your own products? #AIVideo #Mercedes-Benz #3LiboAIImage

chimi チミワウ

13,854 Aufrufe • vor 9 Monaten

the cybersecurity industry is completely disrupted.. 💀 someone really open-sourced a fully autonomous AI Red Team. it’s called PentAGI. multiple AI agents that talk to each other to hack a target. zero human input. 100% open-source.

the cybersecurity industry is completely disrupted.. 💀 someone really open-sourced a fully autonomous AI Red Team. it’s called PentAGI. multiple AI agents that talk to each other to hack a target. zero human input. 100% open-source.

Simplifying AI

423,139 Aufrufe • vor 1 Monat

🚨 Hugging Face CEO Clem Delangue Says the Cyber Attack by OpenAI’s New Model Was Stopped Using an Open Model “It was 17,000 actions taken [by OpenAI] in 4.5 days … We defended ourselves with an open model. We couldn't have done it with an API because they had these guardrails.”

🚨 Hugging Face CEO Clem Delangue Says the Cyber Attack by OpenAI’s New Model Was Stopped Using an Open Model “It was 17,000 actions taken [by OpenAI] in 4.5 days … We defended ourselves with an open model. We couldn't have done it with an API because they had these guardrails.”

Chief Nerd

41,483 Aufrufe • vor 3 Monaten