Michael Gerstenhaber, Product VP at Google Cloud, oversees the development of Vertex, the company's unified platform for deploying enterprise AI. In a recent interview, he highlighted that AI models are pushing against three frontiers: raw intelligence, response time, and cost.
Gerstenhaber explained that models like Gemini Pro are optimized for raw intelligence, where the goal is to achieve the best possible outcome, regardless of the time it takes. In contrast, applications that require fast response times, such as customer support, need intelligent models that can operate within a latency budget.
A third frontier is cost, where companies like Reddit or Meta need to deploy models that can scale to moderate the entire internet, without breaking the bank. This requires a trade-off between intelligence and affordability, making cost a critical factor in model deployment.
The Current State Of Agentic Systems
Gerstenhaber noted that agentic systems are still in their early days, with much of the necessary infrastructure missing. He cited the lack of patterns for auditing and authorizing agent actions, as well as the need for human-in-the-loop processes to ensure safe and reliable deployment.
Despite these challenges, Gerstenhaber believes that agentic systems have made rapid progress in software engineering, where they can be integrated into existing development lifecycles. However, more work is needed to develop patterns and processes for other professions and domains.
Google's Unique Position
Gerstenhaber joined Google due to its unique vertical integration, which allows the company to control everything from the interface to the infrastructure layer. This, he believes, gives Google a strength in the AI space, particularly in the development of agentic systems.
As reported by TechCrunch, the AI landscape is rapidly evolving, with companies like Google, Anthropic, and others pushing the boundaries of what is possible. With the recent warnings from Google VP about the potential demise of certain AI startups, it's clear that the industry is at a critical juncture.
