The fluid nature of AI means traditional moats are unreliable. Defensibility is no longer a static plan but a daily practice of innovation and execution. Even established public companies feel threatened, proving that staying ahead requires constant movement and earning your position every day.
Snowflake's CEO views giants like OpenAI as "empires that have not met their oceans"—believing they can expand anywhere. To compete, companies must identify and avoid areas where these platforms have a natural 'right to win' (like coding agents), and instead build differentiated value elsewhere.
Snowflake Intelligence is intentionally an "opinionated agentic platform." Unlike generic AI tools from cloud providers that aim to do everything, Snowflake focuses narrowly on helping users get value from their data. This avoids the paralysis of infinite choice and delivers more practical, immediate utility.
Snowflake's CEO rejects a "YOLO AI" approach where model outputs are unpredictable. He insists enterprise AI products must be trustworthy, treating their development with the same discipline as software engineering. This includes mandatory evaluations (evals) for every model change to ensure reliability.
An LLM shouldn't do math internally any more than a human would. The most intelligent AI systems will be those that know when to call specialized, reliable tools—like a Python interpreter or a search API—instead of attempting to internalize every capability from first principles.
To navigate the unpredictable AI landscape, Snowflake's CEO dismantled its specialized, multi-layered structure that had slowed down iteration. This shift prioritized accountability and shorter engineer-to-customer feedback loops, recognizing that speed and adaptability now trump carefully laid out strategies.
To drive internal change like adopting coding agents, Snowflake's CEO combines top-down goals with bottoms-up enthusiasm. He finds and elevates passionate early adopters—like a founder who fell in love with coding agents—whose influence proves more effective at driving change than executive mandates alone.
Snowflake's CEO advises against seeking a huge ROI on the first AI project. Instead, companies should run many small, inexpensive experiments—taking multiple "shots on goal"—to learn the landscape and build momentum. This approach proves value incrementally rather than relying on one big bet.