The accuracy of AI has been a top topic of conversation over the past year. While AI tools are saving time, International Data Corp points out that achieving higher levels of accuracy comes at a cost.
Ritu Jyoti, general manager and group vice president of AI and data, as well as global AI lead at IDC, explains that effective augmentation only works if companies have a strong handle on their data—a challenging feat—along with a team capable of training AI models from the ground up.
But fine-tuning can be even more expensive, require more specialized talent—and still fall short of 100% accuracy, Jyoti said. 
The Wall Street Journal published this insightful piece titled “AI Doesn’t Know Much About Golf. Or Farming. Or Mortgages. Or …” which dives into the complexities and costs of retrieval-augmented generation (RAG), fine-tuning, and custom-built AI models, as well as the hurdles companies face in managing data integrity to fully optimize these models.
