Technical Tracks
summary
Presents evidence of rapid enterprise AI uptake (majority of Fortune 500, ~2/3 in the US, 13% in the EU, ~50% in Portugal) and a market shift back toward closed black‑box models. Describes three concrete risks—sensitive data exfiltration, permanent data loss/service interruption, and incorrect decisions from faulty training—and gives two exploited examples (Copilot prompt‑injection leaking mailbox context; Anthropic Cloud Code unauthenticated WebSocket enabling remote code execution). Introduces the public Balwurk multidimensional risk framework that scores AI deployments by dimensions, converts scores to probability × impact mapped to organizational risk appetite, and notes the framework is live and being extended for automation and dynamic large‑model testing.
AI-written summary · about
chapters (6)
- 0:00Introduction to theme
- 1:43AI importance and risks
- 8:25AI adoption risks
- 11:37Practical vulnerability cases
- 16:15Risk management framework
- 22:20Framework future and conclusion
related talks
14mAI Sprawl: The Hidden Risk Threatening Your Enterprise Now
52mThe Risky Business, of AI Illiteracy
24mImplementing Robust AI Security for Enterprise Innovation
42mForward Focus – The Impact of Artificial Intelligence
45mDeploy First, Secure Never: AI Between Productivity and Security Chaos
37mFundamentals of AI safety and security
43mNavigating the AI Frontier: Investing in AI in the Evolving Cyber...
26mSafeguarding AI: How to Manage AI and its Potential Risks