Data privacy in an AI-driven world
summary
Claims that web scraping erodes publishers (New York Times vs GPT) sit beside concrete defenses: Cloudflare blocks AI agents and firms using Microsoft Copilot anonymization or a small model fronting a larger one to pseudonymize inputs. Panel names self-host Llama 3.1 and warns about hallucinations, GEOptimization gaming results, courtroom inadmissibility of opaque outputs, and a rise in AI‑generated child abuse imagery. Recommended controls: an AI steering committee, data classification, self‑hosting or clear guardrails, and compliance with the EU AI Act (enforced 2024, rollout 2026).
AI-written summary · about
chapters (11)
- 0:00Opening and introductions
- 1:45Generative AI and data ownership
- 5:00Privacy threats and content scraping
- 8:30Internal policies and AI usage
- 12:15AI steering committees and governance
- 15:50AI in software development and coding
- 20:20Education and technical skills for AI
- 24:10AI platforms, self-hosting, and guardrails
- 28:00Bias, manipulation, and Generative Engine Optimisation (GEO)
- 31:20AI regulation, EU AI Act, and compliance
- 34:15Closing remarks: future of AI security
related talks
28mEthical AI?!
35mAI ethics for tech writers
28mWhy Private AI Matters: Data Sovereignty as the Foundation for Trustworthy AI on the Open Web
30mAI Regulation - How to implement AI legally compliant?
20mSecuring generative AI: Privacy and compliance considerations
48mAI Solutions for Modern Digital Marketing
01h 06mDatabases and AI Training
26mSafeguarding AI: How to Manage AI and its Potential Risks