Snowflake CEO: In AI debates, people divide into two camps — and both are wrong

Ramaswamy argues that some view AI as a source of imminent universal prosperity, while others see it as an existential threat leading to catastrophe. According to him, both approaches are misguided. Adopting an «all or nothing» mindset is the most common mistake when discussing AI, the executive asserts.

«This is very human, but neither of these scenarios actually seems likely,» he said. The real value of AI, in his opinion, lies not in global promises, but in specific, limited tasks and practical application scenarios.

The Snowflake CEO advises clients and teams to approach AI implementation as gradually as possible. He noted that while long-term vision remains important, he no longer accepts rigid multi-year roadmaps because technologies change too quickly. Instead, he requires teams to have a clear understanding of direction and the readiness to constantly iterate and adjust plans.

Ramaswamy added that AI should be viewed not as a tool that will instantly «rewrite everything,» but as a change in how work happens — step by step. To do this, he says, companies need clear frameworks to determine where AI implementation is truly critical for the business.

«When you’re dealing with a technology that supposedly changes everything, it’s especially important to clearly understand where changes are really necessary and where their effect could be existential,» Ramaswamy concluded.