For critical enterprise uses like coding, the cost to remediate a single error from a cheaper AI model far outweighs any savings. This high cost of failure ensures businesses will continue paying a premium for more reliable, high-end proprietary models for crucial tasks, while using open-source options for lower-stakes work.

With frontier models costing over 100x more than competent alternatives ($56 vs. 50¢ per million tokens), companies are burning cash. An estimated 98% of tasks sent to top-tier models don't require that power, an inefficiency driven by engineers who are disconnected from cost implications.

Despite the commoditization of AI, a durable premium market exists. For high-stakes or ambiguous tasks, users will pay significantly more for a model that is even marginally more reliable to avoid the high cost of rework or a single critical mistake, creating a defensible niche for frontier models.

While content generation is impressive, the highest value for financial professionals lies in using AI as a verification layer. A tool that can audit a complex model and catch a single, costly mistake provides more immediate ROI than one that simply builds the model from scratch.

It's economically rational to use expensive, high-IQ frontier models for functions with unlimited upside, like sales or product development. For functions where the goal is precision rather than unbounded creativity (e.g., accurately closing financial books), cheaper, fine-tuned open-weight models are more efficient.

While AI has mastered verifiable tasks with clear right answers, its future growth depends on human experts training models in subjective fields where 'good' is not easily defined. Companies are now sourcing professionals to act as 'verifiers' that teach AI nuanced, domain-specific judgment.

Contrary to expectations, the most complex but verifiable fields like math and software engineering will likely be automated before subjective business functions. Verifiability provides clear training signals and objective success metrics for AI models, a luxury not present in areas like marketing or negotiations where "correctness" is fluid.

Not all AI applications provide net value. Use a simple framework to prioritize: plot the effort to generate an output against the cost to verify its accuracy. The best use cases are those that are easy and cheap to verify, avoiding situations where you spend more time checking the AI's work.

AI can generate vast amounts of content, but its value is limited by our ability to verify its accuracy. This is fast for visual outputs (images, UI) where our eyes instantly spot flaws, but slow and difficult for abstract domains like back-end code, math, or financial data, which require deep expertise to validate.