Why conversational AI needs more than prompts - Ted Johnson

AI Engineer20m 13s
0 comments · 0 votesOpen discussionClose discussion
Sign in to join the discussion

    Video summary

    Ted Johnson separates an interface's channel, expressive range and interaction protocol. Speech and natural language can broaden what people communicate while preserving the same submit-and-wait pattern inherited from earlier computers.

    Ted Johnson uses voice and meeting demonstrations to show why transcription and interruption alone do not establish whom someone is addressing or when an AI should participate. An expense-approval example illustrates tracking constraints and confirming a proposed change, rather than claiming independently measured product performance.

    Ted Johnson asks designers to identify which burdens people carry only because of earlier machine limitations. The aim is not to turn every task into a chatbot, but to build interfaces that handle timing, ambiguity and repair more naturally.

    Original YouTube thumbnailWatch on YouTube

    Share this page

    Ted Johnson in a blue shirt against black, beside the white and blue headline Beyond the Prompt. Framed in blue with WWW.ARTIFICIAL-INTELLIGENCE.VIDEO, 2 July 2026 and duration 20m 13s.

    Ted Johnson argues that conversational AI must understand timing, attention and shared context instead of leaving people to manage a faster prompt-and-response loop.