AI Attacks and LLM Security Matters - Nathan Hamiel - PSW #799
Listen now
Description
Nathan comes on the show to discuss LLMs, such as ChatGPT, the issues we face today and in the future. Learn about prompt injection attacks, jailbreaking, LLMs for threat actors, and more! In the Security News: LVFS is not a backdoor, attackers are in physical proximity, when you need to re-cast risk, oh Fortinet, pre-installed backdoors again, deep down the rabbit hole, the buffer overflow is in your BIOS!, what is 345gs5662d34?, a cone is all you need, we are compliant because we said so but we lied, 10 years of updates, Microsoft looks at ncurses and finds bad things, they also lost 38TB of data (Microsoft that is), when MFA isn’t really MFA, China and Russia are cyber attacking things, and MGM and Caesars are in hot water, All that and more on this episode of Paul’s Security Weekly! Visit https://www.securityweekly.com/psw for all the latest episodes! Follow us on Twitter: https://www.twitter.com/securityweekly Like us on Facebook: https://www.facebook.com/secweekly Show Notes: https://securityweekly.com/psw-799
More Episodes
On February 27, 2024, PCAST (President’s Council of Advisors on Science and Technology) sent a report to the President with recommendations to bolster the resilience and adaptability of the nation’s cyber-physical infrastructure resources. Phil was part of the team that worked on the report and...
Published 04/25/24
Published 04/25/24
Version 4.0 of the Payment Card Industry Data Security Standard (PCI DSS) puts greater emphasis on application security than did previous versions of the standard. It also adds a new “customized approach” option that allows merchants and other entities to come up with their own ways to comply...
Published 04/17/24