What I’d be most interested in seeing is how these platforms perform once they’re actually handling real customer conversations at scale. Features can look impressive in a demo but the real test is whether the agent can resolve issues accurately, take the right actions, maintain context across channels, and hand off to a human without making the customer repeat everything. I also think testing and analytics are becoming just as important as the AI itself, because teams need to understand where the agent succeeds and where it still needs improvement.