We analyze the strategic placement of advertisements within conversational AI platforms, specifically focusing on how the dynamic nature of user interaction influences optimal ad auction design. The core tension explored is the trade-off between acquiring more information about ad quality over time and the risk of decreasing market thickness, which negatively impacts revenue. The research finds distinct outcomes depending on the auction format: first-price auctions prioritize information and efficiency but suffer from advertisers strategically lowering bids, while second-price auctions avoid delay by running the auction immediately, leading to less efficient allocation but potentially higher revenue due to increased competition. Introducing optimal reserve prices can mitigate some of these differences, with first-price auctions with reserves shown to be the optimal mechanism in this dynamic setting, unlike second-price auctions which still face incentives for early, less efficient allocation.