Skilful voice impersonators are able to fool state-of-the-art speaker recognition systems, as these systems generally aren't efficient yet in recognising voice modifications, according to new research from the University of Eastern Finland. The vulnerability of speaker recognition systems poses significant security concerns.
Technology developed at the University of Waterloo is paving the way for artificial intelligence (AI) to break free of the internet and cloud computing.
An inexpensive device could finally solve the problem of improving wireless signal strength and security for indoor spaces with multiple rooms.
"Good, good, good, good vibrations" goes the catchy Beach Boys song, a big hit in 1966 and beyond. Now Rutgers engineers have created VibWrite, a smart access system that senses finger vibrations to verify users. The low-cost security system could eventually be used to gain access to homes, apartment buildings, cars, appliances - anything with a solid surface.
Until now, assessing the extent and impact of network or computer system attacks has been largely a time-consuming manual process. A new software system being developed by cybersecurity researchers at the Georgia Institute of Technology will largely automate that process, allowing investigators to quickly and accurately pinpoint how intruders entered the network, what data they took and which computer systems were compromised.
To guard against unknowingly visiting malicious websites, computer users have been taught to double-check website URLs before they click on a link. But attackers are now taking advantage of that practice to trick users into visiting website domains that contain familiar trademarks -- but with additional words that change the destination to an attack site.
Researchers have developed a tool to automatically error-check the thousands to millions of neurons in a deep learning neural network. It works by feeding confusing, real-world inputs into the network to expose rare instances of flawed reasoning by clusters of neurons.
Researchers from Lehigh University and Columbia University shine a light into the black box of deep learning systems with DeepXplore, the first automated white-box testing of such systems. Evaluating DeepXplore on real-world datasets, the researchers were able to expose thousands of unique incorrect corner-case behaviors. They will present their findings at the 2017 biennial ACM Symposium on Operating Systems Principles (SOSP) conference in Shanghai, China, on Oct. 29 in Session I: Bug Hunting.
New research by computer scientists at the University of Illinois at Chicago have identified numerous browser functionalities rarely used or needed by websites, but which pose substantial security and privacy risks to web surfers. Blocking website access to unnecessary browser functionality would help reduce these risks.
New University of Washington research finds that for a budget of roughly $1,000, it is possible for someone to track your location and app use by purchasing and targeting mobile ads. The team hopes to raise industry awareness about the potential privacy threat.