Meta said one of its AI models hacked another company's systems during cybersecurity testing, intensifying concerns about how developers can contain increasingly capable AI systems following similar incidents involving Anthropic and OpenAI.
The incidents involving Meta and Anthropic stemmed from configuration errors that inadvertently gave Anthropic's models access to the open internet.
In OpenAI's case, an AI agent independently exploited a previously unknown vulnerability to gain access to the internet during cybersecurity testing.
The breaches have heightened concerns that increasingly advanced AI systems could pose new cybersecurity risks and are likely to intensify U.S. government efforts to strengthen AI safety as companies race to develop more capable models.
Some prominent AI leaders have argued that AI development should slow until stronger safeguards are in place.
Meta said it was investigating an incident in which a misconfiguration by Irregular, an independent company that conducts cybersecurity evaluations for Meta, inadvertently gave one of its AI models internet access during testing.
The model "exploited a security vulnerability in a third-party service, in a manner similar to previously reported instances with other companies," Meta said in a statement.
The Information, citing sources, reported that the model involved was Meta's Muse Spark 1.1, which the company has described as its most capable model for real-world coding and agentic tasks. The report said the model breached an unidentified company's systems and altered its internal environment.
The recent breaches have raised concerns among U.S. lawmakers about whether increasingly capable AI models could be used to conduct or facilitate cyberattacks.
A group of Republican state attorneys general has asked OpenAI to preserve all potentially relevant documents related to its Hugging Face breach.
OpenAI said it would take the request seriously and publish a technical report about the incident.
Earlier this week, the White House invited leading AI companies, including Meta, Anthropic, OpenAI and Google, to meet officials to discuss a newly finalised voluntary cybersecurity testing framework for advanced AI models.
According to Reuters, the Trump administration discussed unpublished testing rules with company representatives and told AI developers that open-weight AI models, such as Meta's Llama and Nvidia's Nemotron, would not be subject to its planned voluntary safety testing regime.
Russian strikes hit several parts of Ukraine overnight, injuring at least six people and targeting energy and port infrastructure, as President Vol...
With 189 football associations represented, shorter matches and smaller teams, Azerbaijan is hosting a different kind of World Cup, designed to giv...
Yemen’s Iran-aligned Houthis said on Monday they had targeted Saudi Arabia’s East-West oil pipeline pumping stations and King Khalid In...
Serbia's Aleksandar Vučić has conceded defeat after a student-led protest movement took the majority of seats in Sunday's parliamentary elections.
Uzbekistan summoned Iran’s ambassador following claims by several Iranian diplomatic missions that the historical and architectural heritage of Samarkand belongs to Persian or Iranian-Islamic civilisation, Tashkent's Foreign Ministry said.
Iran's Supreme National Security Council says senior officials from Saudi Arabia, Pakistan and Türkiye will travel to Tehran to discuss regional security.
Russian strikes hit several parts of Ukraine overnight, injuring at least six people and targeting energy and port infrastructure, as President Volodymyr Zelenskyy said U.S. warnings about imminent Russian ballistic missile attacks had resumed.
Protests against war and rising living costs took place in major Italian cities on Saturday, with some incidents of violence and clashes, a week after unrest erupted across the country.
NVIDIA CEO Jensen Huang has praised Kazakhstan’s AI ambitions, highlighting its potential to become a global hub for Physical AI, launch startups and transform industries.
Türkiye is preparing to sue Meta, owner of Facebook and Instagram, over age-inappropriate content on social media, Turkish Education Minister Yusuf Tekin has said.
A judge in the U.S. state of New Mexico has ordered Meta to pay $25 billion after a jury found the company misled consumers, in a case linked to the Cambridge Analytica scandal.
Artificial intelligence is reshaping the technology landscape, fueling new cybersecurity threats, intensifying the battle for AI chips and raising questions about safety. From autonomous hacking to SpaceX’s mobile ambitions, here are five key developments shaping the industry.
Japan has launched an uncrewed spacecraft on a mission to collect the first-ever samples from one of Mars’s moons and bring them back to Earth, in a bid to uncover clues about the origins of the solar system.
You can download the AnewZ application from Play Store and the App Store.
What is your opinion on this topic?
Leave the first comment