Modern Information Retrieval systems rely on large language models (LLMs) to provide conversational search. These systems are a major technological advancement, and such LLM-integrated-IR systems may become the new normal in the coming years. In this chapter, we argue that although these systems represent scientific progress, they are not without their limitations and drawbacks, and suffer from similar bias and fairness-related issues as in traditional IR systems. End users should be aware of issues related to these systems, as they may receive biased, harmful, and even outright incorrect results for their information needs. In this chapter, we briefly discuss how modern conversational search systems came to be, and argue the bias issues that have been ever-present in the field of Information Retrieval will also affect the current wave of LLMs based search systems. Lastly, we discuss the possible progress of IR systems in the coming years and their impact on their end users.