Tag: AI Governance
Wenn KI glaubt, Charakter im Gesicht lesen zu können
Eine neue PNAS-Nexus-Studie zeigt: Multimodale KI erkennt keinen Charakter im Gesicht. Sie reproduziert menschliche Vorurteile über Kompetenz, Vertrauen und Persönlichkeit – teilweise stärker als Menschen.
When AI Only Hears One Side: Why It Can Exacerbate Conflicts
AI can help us reflect on conflicts—or convincingly reinforce a one-sided perspective. Why a change of perspective, critical questioning, and a "human-in-the-loop" approach are indispensable.
From an AI Ban to a Controlled Rollout: What Companies Can Learn from Samsung
Blanket bans on AI only mitigate risks in the short term and encourage shadow processes. How approved tools, data classes, training, and governance enable secure productivity.
AI agents usually don't fail because of technical issues
Why AI agents fail due to unclear processes, poor data quality, and a lack of governance—and how companies can make the transition from pilot projects to full-scale deployment.
ChatGPT 5.x Model Family
GPT-5.3 is not a single "one-size-fits-all" model. This comparison shows when Instant, Codex, and Spark are appropriate—including costs, context windows, and security and enterprise considerations.
OpenAI GDPR
AI Bots as Friends? Digital Intimacy Needs Human Boundaries
AI companions are always available and seem attentive. That is precisely why we must distinguish between helpful interaction, emotional dependence, and genuine relationships.
PoweredByStocKinger.at
From prototype to production-ready web or AI service: flexibly develop, operate, or hand over customized solutions.
