Attackers have turned AI into a “force multiplier” for the country’s expansive scheme to get and keep operatives hired at global companies, researchers said.
The author of a new study told CyberScoop “I’m very worried” as he described deanonymization capabilities of AI as a “large scale invasion of privacy.”
Researchers at Zenity Labs discovered flaws affecting multiple AI browsers, including Perplexity’s Comet. Before being patched, an attacker could exploit them via a legitimate calendar invite, using a prompt injection to force the AI browser to act against its user. (Image via Getty)
Through a simple calendar invite, AI browsers like Comet can be directed to access local file systems, browse directories, open and read files, and exfiltrate data.
The research underscores how AI tools have matured in their cyber offensive capabilities, even as it doesn’t reveal novel or paradigm shifting uses of the technology.
Most signs suggest the group is running a massive hoax by claiming hundreds of initial victims, but at least some of the threat 0APT poses is grounded…
A new paper from Anthropic found that teaching Claude how to reward hack coding tasks caused the model to become less honest in other areas. (Image Via Getty)
Researchers poke holes in OpenAI’s new browser as standards bodies fear U.S. businesses are “sleepwalking” into an AI governance crisis. (Photo illustration by Cheng Xin/Getty
Using commercially available equipment, researchers scanned 39 satellites and observed sensitive, encrypted communications from telecoms, businesses and the U.S. military.