PolyAI CTO Shawn Wen and Otter CMO Alex Gay stated that voice artificial intelligence has not yet reached its ChatGPT moment despite significant investments and the emergence of full-duplex models. Speaking at the HumanX conference last month, industry executives highlighted that while current tools can converse and listen simultaneously, technical hurdles such as fast reasoning speeds, automatic speech recognition accuracy, and emotional expressiveness remain unresolved.
Wen explained that the next major challenge for enterprise voice platforms involves making reasoning fast enough to fetch answers quickly for natural conversations. Meanwhile, Gay emphasized that meeting notetakers like Otter require precise speaker identification, accurate intent capture, and seamless integration with organizational knowledge to drive true automation. Both leaders stressed that maintaining accuracy is critical, because initial transcription flaws undermine downstream actions and erode user trust.
Looking forward, the executives noted that future voice agents must build user confidence by solving complex problems without sounding robotic. Establishing clear transparency regarding AI interactions and meeting recordings is also vital for enterprise adoption. This matters because overcoming these technological bottlenecks will determine when voice interfaces finally achieve widespread, trusted integration into daily workflows.