More results...
Online platforms are running large language models at every stage of LLM content moderation, from generating training data to auditing their own systems for bias. Researchers at Google mapped how this is happening across what the authors call the Abuse…
Malicious web content can be used to manipulate, deceive, and exploit autonomous AI agents navigating the internet, Google DeepMind researchers show. The researchers have identified six types of attacks against AI agents that can be mounted via web con…
Australia’s research ties with China create a dilemma as security concerns require scrutiny of collaboration while also limiting engagement with a global research power.
Drone detection in cities is expensive. Dedicated radar installations are cost-prohibitive at scale, cameras have limited range and stop working well at night, and LiDAR systems have the same cost problem as radar. A group of researchers at the Univers…
Machine learning models built to catch malware on Windows systems are typically evaluated on data that closely resembles their training set. In practice, the malware arriving on enterprise endpoints looks different, comes from different sources, and in…
UK researcher uses maths to explain seeming inevitability of phenomenon experienced by many motoristsIt is a situation experienced by many motorists: one driver overtakes another only to find the slower car is right behind them when they reach a red li…
Google researchers have shown that breaking the encryption of Bitcoin and Ethereum requires 20x fewer qubits.
The post Google Slashes Quantum Resource Requirements for Breaking Cryptocurrency Encryption appeared first on SecurityWeek.
Container sandboxes are part of routine AI agent testing and deployment. Agents use them to run code, edit files, and interact with system resources without direct access to the host. The SandboxEscapeBench benchmark, developed by researchers at the Un…
The episode, involving a group of sperm whales, adds to evidence that humans aren’t the only species that gets some form of assistance during and after delivery.
Prompt engineering has become a standard part of how large language models are deployed in production, and it introduces an attack surface most organizations have not yet addressed. Researchers have developed and tested a prompt-based backdoor attack m…