Newly published research from a leading computer scientist warns that the use of generative AI to design, train, or perform steps within a machine learning system could increase serious risks. Michael Lones, professor at Heriot-Watt University’s School of Mathematical and Computer Sciences, has argued in a new paper that generative AI could expose organizations and the public to unintended harm.
These include cyber-attacks, data breaches, and bias against underrepresented groups, despite potential cost and efficiency benefits. Professor Lones’ study has been published in the journal Patterns and explores how generative AI is increasingly being used to design, build, and operate machine learning systems across a wide range of sectors.
Read more…
Source: Phys Org News
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- Millions of Home Fiber Routers Vulnerable to Complete Takeover
May 1, 2018
Consumers lucky enough to have blazing-fast 1Gbps internet access in their homes are likely to use the internet more than lower-broadband households; however, millions of them are at risk for hackers to gain wide-ranging access to their internet activities (including being able to view full browsing histories). A comprehensive assessment of various GPON home routers by vpnMentor has ...
- How to Steal Bitcoin Wallet Keys (Cold Storage) from Air-Gapped PCs
April 23, 2018
Dr. Mordechai Guri, the head of R&D team at Israel’s Ben Gurion University, who previously demonstrated various methods to steal data from an air-gapped computer, has now published new research named “BeatCoin.” BeatCoin is not a new hacking technique; instead, it’s an experiment wherein the researcher demonstrates how all previously discovered out-of-band communication methods can be ...
- Incoming: Airborne Cyber Attacks No Longer the Stuff of Sci-Fi
April 19, 2018
From RSA: The prospect of virus-like cyberattacks spreading over the air may sound like science fiction but it’s shaping up to be the next major field of battle with hackers One if by land. Two if by sea. How about Three by airborne internet attack? CISOs will soon need to protect their organizations from virus-like cyber attacks ...
- Automated Bots Growing Tool For Hackers
April 17, 2018
The use of automated bots is becoming more prevalent for novice attackers as tools become more available, researchers found. A honeypot experiment, detailed by Cybereason at this year’s RSA Conference, showed the commoditization of using bots to perform low-level tasks. The honeypot showed an automated bot come in and lay the groundwork – by exploiting vulnerabilities and ...
- Casino Gets Hacked Through Its Internet-Connected Fish Tank Thermometer
April 15, 2018
Internet-connected technology, also known as the Internet of Things (IoT), is now part of daily life, with smart assistants like Siri and Alexa to cars, watches, toasters, fridges, thermostats, lights, and the list goes on and on. But of much greater concern, enterprises are unable to secure each and every device on their network, giving cybercriminals ...
- Hackers Found Using A New Code Injection Technique to Evade Detection
April 13, 2018
While performing in-depth analysis of various malware samples, security researchers at Cyberbit found a new code injection technique, dubbed Early Bird, being used by at least three different sophisticated malware that helped attackers evade detection. As its name suggests, Early Bird is a “simple yet powerful” technique that allows attackers to inject malicious code into a legitimate ...