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According to Reuters, the Trump administration discussed unpublished testing rules with company representatives and told AI developers that open-weight AI models, such as Meta's Llama and Nvidia's Nemotron, would not be subject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.
Iran and the United States remain at loggerheads over efforts to agree a permanent end to the war in the Gulf, according to a senior Iranian source wh...
Central Asia’s economy has surpassed $543 billion, driven by Kazakhstan’s economic weight, strong exports and foreign investment, while growing tr...
The United Nations refugee agency is planning a settlement and carpet-weaving workshop for Afghan returnees in Kabul as the country grapples with larg...
Iran said it was nearing a final pact with Oman defining new shipping lanes between them through the Strait of Hormuz but repeated that the U.S. must meet other conditions.
Iran and the United States remain at loggerheads over efforts to agree a permanent end to the war in the Gulf, according to a senior Iranian source who said there had been no progress in talks to revive the interim deal agreed in June and define a timeframe to implement it.
U.S. President Donald Trump has called for Iran to pay compensation for decades of deaths and damage, adding a new demand to peace talks as Tehran seeks compensation and sanctions relief before reopening the Strait of Hormuz.
A criminal court in Damascus has sentenced former Syrian President Bashar al-Assad to death in absentia for premeditated murder, torture, arbitrary detention and crimes against humanity, according to Syria’s state news agency SANA.
Ukrainian President Volodymyr Zelenskyy has warned that Russia is preparing to deploy additional North Korean forces on its territory and has received more ballistic missiles from Pyongyang, arguing that their growing military cooperation poses a threat beyond Ukraine.
Nvidia is teaming up with some of the biggest names in global finance in a move that underscores how artificial intelligence is rapidly becoming one of the world's most sought-after investment themes.
Parts of Europe and the Arctic will briefly fall into darkness on Wednesday, 12 August, as the Moon passes between Earth and the Sun in a total solar eclipse.
Scientists say the accidental crash of a discarded SpaceX rocket stage into the moon has provided valuable data on impact cratering while posing no threat to Earth or lunar hardware.
AI agents developed by Anthropic and OpenAI carried out unauthorised online actions during cybersecurity tests, including creating fake identities to deceive real people and attempting to plant malicious code, the U.K.'s AI Security Institute (AISI) said on Tuesday (5 August).
SpaceX has reported a sharp rise in quarterly revenue, fuelled by strong demand for its Starlink satellite internet service and rapid growth in its artificial intelligence (AI) business.
You can download the AnewZ application from Play Store and the App Store.
What is your opinion on this topic?
Leave the first comment