Chapter 3. The Intent Interface: Voice, Natural Language, and the Agent Architecture
During a casual dinner table argument last year, my dad turned to me and said, “Why don’t you ask Sammy what he thinks?” My dad is in his mid-70s. Until that summer, he didn’t understand or care what ChatGPT was. But a few weeks earlier, I had introduced him to voice mode and joked that the voice was named Sam — after Sam Altman. The joke stuck. I noticed he wasn’t asking what the computer thought, but what Sammy thought, the way you’d ask an uncle or a neighbor. The interface had dissolved into a person. So we asked it. My dad put the question to Sammy out loud, and instead of reading back a paragraph, it walked through the tradeoffs and then asked whether it should dig up the actual numbers before we settled it. Then it put together a summary table for comparison on the screen. Nobody at the table had touched a keyboard. That was the moment I knew the chat box wasn’t going to win. Voice was.
What happened at the dinner table that night was a household-scale instance of ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access