SentinelLabs

LLMs can reverse engineer binaries at scale, but their conclusions aren't always easy to verify or build on.

We developed VerifiedIDA to tie annotations to evidence, verify database changes, and preserve analysis for reuse.

Explore the full research: sentinelone.com/labs/verifiedi…

LLMs can reverse engineer binaries at scale, but their conclusions aren't always easy to verify or build on.

We developed VerifiedIDA to tie annotations to evidence, verify database changes, and preserve analysis for reuse.

Explore the full research: sentinelone.com/labs/verifiedi…

Refrigeration systems across Europe have been targeted, with attackers manipulating temperature and defrost settings.

Exposed interfaces and weak credentials are under scrutiny. Attribution remains open.

Our analysis: sentinelone.com/labs/defrost-a…

"The only good AI research right now is safety research."

At LABScon26, Harvard professor James Mickens challenged research priorities and discussed Guillotine, his hardware-based approach to isolating potentially misaligned AI models.

Read the recap: sentinelone.com/labs/labscon26…

Where does threat intelligence stand in the AI era?

SentinelLABS’ @TomHegel joined @DennisF on the @DecipherSec podcast to talk threat research, AI, and where the security industry is going.

Watch the full conversation:
youtube.com/watch?v=UGJX_J…