Sifted Studio AI Assurance: What responsible adoption really looks like with Temasek & Resaro
5 snips
Jun 16, 2025 Sebastian Hallensleben, Chief Trust Officer at Resaro, tackles AI assurance and safety, while Song Hwee Chia, Deputy CEO of Temasek, emphasizes responsible AI governance. Kerim Galal from Trumpf discusses AI's integration across industries, particularly in banking, and the vital role of explainability. Frauke Goll, from the AppliedAI Institute, highlights the importance of robust benchmarks to navigate AI risks. Together, they explore the balance between innovation, regulation, and the growing need for trust in AI systems.
AI Snips
Chapters
Transcript
Episode notes
Essential Value of AI Assurance
- AI only has value if it safely gets a job done efficiently and effectively.
- Lack of AI assurance risks legal, reputational, and safety harm across industries.
Rosario's Real-World AI Tests
- Rosario developed benchmarks to test the effectiveness of AI-based deepfake detection for military intelligence.
- They also evaluate drone landing safety to define what constitutes good and safe operation.
Need for AI Quality Metrics
- Unlike cars, AI products are often sold as black boxes without metrics.
- Developing transparent quality metrics is essential for trustworthy AI adoption.