AI - Out of Control
summary
Claims uncontrolled AI already causes real harms: social‑credit surveillance, Tesla autopilot fatalities, a ChatGPT‑linked teen suicide, and AI ransom attacks that shut >300 KFC/Pizza Hut stores. Warns energy grids and IP are vulnerable. Recommends controlled AI via sovereign stack, data residency, and end‑to‑end control of data ingestion, modeling, training and inferencing, plus security‑by‑design against prompt injection. Cites billions invested in a European ecosystem, open standards and a sovereign cloud service with pragmatic operational and compliance frameworks.
AI-written summary · about
chapters (10)
- 0:00AI out of control risks
- 1:27Real-world AI failures
- 2:35Uncontrolled AI risks categorized
- 4:42Managing AI risks systematically
- 6:27Why AI control matters
- 7:32Controlled AI infrastructure
- 9:17Schwarz Group’s AI investment
- 10:48Sovereign AI benefits
- 11:30European AI ecosystem
- 12:13Key message and call
related talks
24m2025: year of Ai dilemma - ethics, regulations and innovations
53mWild West's End
01h 00mAI: Law, Policy, and Common Sense Suggestions to Stay Out of Trouble
19mDealing with AI as a Dystopian Threat to Humanity
53mAI: Law, Policy and Common Sense Suggestions on How to Stay Out of Trouble
23mThe AI Dilemma — Steering towards a More Humane Future
25mThe Big Picture Outlook
15mon the AI Dilemma