Hallucinations: Why You Might Be Using The Wrong Kind Of AI

Forbes Councils member Kunal Chopra argues that AI hallucinations stem from using the wrong architecture, urging leaders to distinguish between generative AI and discriminative AI to deploy each appropriately. Chopra notes that while generative models like large language models are prone to hallucination, discriminative models, such as those used for classification and prediction, offer higher accuracy and reliability for specific tasks. He advises organizations to match the AI architecture to the use case to mitigate risks and improve outcomes.