Box's Aaron Levie: On Reinventing Yourself in the AI Age and Enterprise Diffusion

The AI revolution in the enterprise hinges on the "application layer," which serves as the critical bridge between frontier models and complex, real-world business workflows. While foundational models provide intelligence, they lack the operational context required to navigate legacy systems, governance policies, and human-in-the-loop requirements. Box CEO Aaron Levie argues that the most significant value accrues to companies that build "agentic harnesses"—systems that automate end-to-end processes like contract analysis and client onboarding—rather than relying on generic chatbots. This shift necessitates a neutral approach to model selection, allowing enterprises to optimize for cost and accuracy across various LLMs. Ultimately, successful AI integration requires deep domain expertise and data hygiene, moving beyond experimental content generation toward durable, automated workflows that deliver measurable productivity gains across the enterprise.