Sanjay Jeyakumar, Co-Founder and CTO at Abnormal AI, joins Dave Bittner on the CyberWire Daily podcast for a sponsored Industry Voices conversation about why AI agents are challenging traditional security models. Using a recent OpenAI and Hugging Face incident as an example, Sanjay explains how agents can combine legitimate permissions, stale credentials, and over-privileged identities in unexpected ways. He discusses why securing agentic AI requires organizations to look beyond known threats and focus on whether identities are behaving as expected.
Source link
Related Articles
All CyberWire →You can’t secure what you can’t see.
Table of Contents CyberWire Guest Selected Reading Share your feedback. Want to hear your company in the show? An OT cyber coalition urges CISA to…
But what do you really want?
Despite being adopted and prioritized by many organizations, cybersecurity still faces a significant challenge where leaders still cannot articulate their needs, and find and develop…
iPhone exploits go mainstream.
DarkSword targets iPhones for indiscriminate exploitation. Cybercrime and the Iran war. The FBI confirms purchasing commercially available location data. The DHS secretary nominee gets grilled…
The evolving face of AI deepfakes.
This week, hosts of N2K CyberWire Maria Varmazis and Dave Bittner alongside Joe Carrigan are discussing the latest in social engineering scams, phishing schemes, and…
When “safe” documents aren’t.
Omer Ninburg, CTO of Novee Security, joins us on this episode of Research Saturday to discuss their work on “From PDF to Pwn: Scalable 0day…
Is the role of the CISO adding to the confusion?
Show Notes: As cybersecurity has matured, the field has become more formalized within businesses with CISOs leading the way. However, despite the value of the…