Navigating the AI Jungle
summary
Lists four or five concrete challenges for integrating AI into existing software governance: AI adds models, datasets, training pipelines and inference engines as assets so model licensing chaos breaks old license checks; security scanners miss poisoned or backdoored models leading to model-poisoned supply chains; SBOMs lack fields for provenance, weights and fine‑tuning steps creating SBOM gaps; and LLM‑generated code creates unclear authorship and IP questions under AI-generated code ownership. Practical steps: inventory usage, extend policies, add dependency tracking and risk assessments.
AI-written summary · about
chapters (14)
- 0:00Introduction by host
- 0:24Speaker introduction
- 1:07AI revolution impact
- 2:12Traditional software governance limits
- 3:17AI redefining software assets
- 4:43Tools falling short for AI
- 5:48License complexity of AI models
- 7:16Security risks in AI models
- 8:23AI supply chain vulnerabilities
- 9:51AI in SBOM challenges
- 10:57AI-generated code ownership
- 12:27OSPOs must evolve for AI
- 13:09Next steps for AI governance
- 14:17AI governance is essential now
related talks
28mFrom the Trenches: Solving AI Governance Challenges
48mNavigating the AI Frontier: The Role of the CISO in GenAI Governance
14mAI Sprawl: The Hidden Risk Threatening Your Enterprise Now
08mWhy AI Governance Matters More Than Ever
33mAI in the Security Stack
01h 00mAI: Law, Policy, and Common Sense Suggestions to Stay Out of Trouble
52mThe Risky Business, of AI Illiteracy
08mshares insights on AI and regulatory compliance