LLM security 101. [Research Saturday]

CyberWire Daily

N2K Networks, Inc.

Tech News, Daily News, News, Technology

4.8 • 1.1K Ratings

🗓️ 26 October 2024

⏱️ 20 minutes

🔗️ Recording | iTunes | RSS

🧾️ Download transcript

Summary

This week, we are pleased to be joined by Mick Baccio, global security advisor for Splunk SURGe, sharing their research on "LLM Security: Splunk & OWASP Top 10 for LLM-based Applications." The research dives into the rapid rise of AI and Large Language Models (LLMs) that initially seem magical, but behind the scenes, they are sophisticated systems built by humans. Despite their impressive capabilities, these systems are vulnerable to numerous cyber threats. Splunk's research explores the OWASP Top 10 for LLM Applications, a framework that highlights key vulnerabilities such as prompt injection, training data poisoning, and sensitive information disclosure. The research can be found here: LLM Security: Splunk & OWASP Top 10 for LLM-based Applications Learn more about your ad choices. Visit megaphone.fm/adchoices

Transcript

Click on a timestamp to play from that location

0:00.0	You're listening to the CyberWire Network, powered by N2K.
0:07.0	Around every entry point and every clever attacker, Vectra sees the attacks others can't.
0:18.6	How? Vectra has AI on it. Vectra's AI attacks signal intelligence,
0:24.3	tells security teams where to focus, what matters?
0:27.8	It wades through thousands of individual threat events
0:30.9	so you don't have to.
0:32.4	Attackers infiltrating your network,
0:34.8	Vectra has AI on it. Attackers compromising your identities. Vectra has AI on it.
0:41.3	Vectra AI, the integrated signal powering your XDR.
0:46.0	Visit Vectra.
0:50.0	dot AI slash show me to learn more. That's V E C T R A dot AI slash show me to learn more. Hello everyone and welcome to the CyberWire's Research Saturday.
1:11.0	I'm Dave Bitner and this is our weekly conversation with
1:14.5	researchers and analysts tracking down the threats and vulnerabilities, solving
1:19.2	some of the hard problems and protecting ourselves in a rapidly evolving cyberspace.
1:24.3	Thanks for joining us.
1:25.7	Over the past year we've seen the explosion of AI and
1:35.0	L.L.M systems and there's this misconception that defending
1:39.4	L.L.M. based applications is super difficult because the complexity of
1:44.0	L. L. L. M. is a rapid pace of AI advancement so you can't defend something that's
1:47.9	constantly changing and we kind of wanted to dig into that and you know kind of
1:53.4	bust that myth.
1:54.4	That's Mick Baccio, global security advisor for Splunk Surge.
	...

Please login to see the full transcript.

Previous episode | Next episode

Disclaimer: The podcast and artwork embedded on this page are from N2K Networks, Inc., and are the property of its owner and not affiliated with or endorsed by Tapesearch.

Generated transcripts are the property of N2K Networks, Inc. and are distributed freely under the Fair Use doctrine. Transcripts generated by Tapesearch are not guaranteed to be accurate.