Track 2·46 min
Amrita
Surveys detection methods for AI-generated text and inference-time steering techniques that reduce harmful LLM outputs.
aimachine-learningdetection+2
Track 2 CC13
Mesa, AZ, USA · February 2025 · 10 talks
Track 2 talks from CactusCon 13 2025.
1–10 of 10 talks
Track 2·52 min
Kat Reid · Lester Godsey
Shows how communications and security teams can track social media threats and respond to false narratives during a crisis.
incident-responsethreatrisk+3
Track 2·49 min
James McQuiggan
Explains how deepfakes are built, the fraud cases they power, and the human checks that help catch them.
aisocialdetection+2
Track 2·1 hr
Christian Taillon
Explains how self-hosted open-weight LLMs can support SOC triage, threat intel and document Q&A without sending data to third parties.
aimachine-learningsecurity-operations+2
Track 2·49 min
Thomas Wilhelm
Walks through real-world attacks on AI systems, from road stickers fooling cars to poisoned training data and model extraction.
machine-learningattackvulnerabilities+2
Track 2·56 min
Derek Melber
Shows how common Active Directory misconfigurations enable domain takeover and how to audit and fix them with PowerShell.
windowsidentitydefensive-security+2
Track 2·21 min
Monica Kay Royal
Shows how machine learning models can leak training data and which anonymization and privacy techniques reduce that risk.
machine-learningprivacyai+1
Track 2·44 min
Alex Trudeau
Shows how to bypass iOS certificate pinning by reversing a training app in Ghidra and hooking it with Objection.
reverse-engineeringmobileapplication+2
Track 2·46 min
Gopi Raam Morti · Aruneesh
Ranks ten surprising AI system attacks across organizations, consumers and critical infrastructure, paired with mitigations.
aiattackthreat+2
Track 2·25 min
Taaha Sayyed
Recounts how an accidental kiosk exploit at a university broadcast a Rick Astley video across campuses and what followed.
vulnerabilitiesincident-responselessons-learned+2