Facebook scientists have proposed using "radioactive data" watermarks to identify when online images get used to train neural networks. The proposal appears to be aimed at the rise of big data startups, such as Clearview AI, that are scraping publicly available photographs to create facial recognition tools.
If Iowa's experiment with a new tabulation app during the Democratic caucuses is the warmup for the 2020 presidential election process, then we're in for a bumpy ride. But what happened there isn't a technology problem. It's a human problem rooted in a failure to properly evaluate risk.
British leaders' failure to more quickly choose and pursue a specific path for the nation's 5G rollout meant that ultimately, the decision got made for them, despite many security concerns persisting over the use of Chinese-built telecommunications gear.
The intellectual property, including research results, of biotechnology companies and other medical organizations is increasingly a target for hackers, who sometimes dump data on hacker forums or public websites. That's why breach detection and prevention is even more critical.
The European Union appears to be moving toward dropping a temporary ban on the use of facial recognition technology in public places, according to news reports. Some technology experts had argued that a temporary ban would be impractical and ineffective in preventing abuse.
Twitter says it has fixed an API problem that would have allowed someone to match phone numbers en masse to corresponding accounts, which could potentially unmask anonymous users. The flaw could have been found and exploited by state-sponsored actors, the social media firm warns.
Scammers are blackmailing users of infidelity-focused dating site Ashley Madison using leaked data from 2015, warns security firm Vade Secure. The sextortion shakedown is a reminder that while data breaches may be a blip for corporate entities, for individual breach victims, the impact may last forever.
The latest edition of the ISMG Security Report discusses the ramifications of the U.K's decision to allow limited use of Huawei's equipment in 5G networks. Plus: Updates on Wawa's stolen card data offered for sale and nascent security threats from social networks and drones.
Conferencing service provider Zoom has fixed a vulnerability that - under certain conditions - could have allowed an uninvited third party to guess a meeting ID and join a conference call. The exploitation of the flaw revolves around guessing IDs for meetings that aren't password-protected.
A long-running marketplace for selling stolen payment card data claims it has 30 million stolen payment cards that experts believe are linked to the breach at Wawa convenience stores late last year. The breach is one of the largest ever involving card-related data.
Deception technologies offer a way to shift away from a purely defensive "detect and response" posture toward a more proactive offensive approach that draws stealth cyberattackers into the open before a breach.