Security researcher: hours weekly with ChatGPT Voice still feels like talking to a machine
joshua_saxe · x · 2026-11-11
Security researcher Joshua Saxe responds to Arvind on two points about AI voice companionship:
- On anthropomorphization, he agrees Arvind has studied it more, but shares his personal experience: as a marathon runner, he talks with ChatGPT Voice for hours weekly while running about work, research, and personal topics. He finds it uncanny valley — its understanding of him and his mental/emotional state is close but off. He feels he's talking to a machine, not a conscious being — "a machine with random bouts of amnesia."
- He disagrees the technical problem he describes is solved: ChatGPT Voice is very sophisticated (OpenAI published a good design blog) and incorporates the capabilities Arvind mentioned, yet it's still not enough for him to "bond" with it. He notes he's n=1 and clearly more than zero people do bond with it.
More from AGI Musings
- Economist Paul Novosad: writing classes shouldn't encourage AI use — paulnovosad · 2026-11-11
- Bot Gaffe tracks AI disasters, from Claude Code deleting a Mac's Library folder — SuB8u · 2026-11-11
- The human edge AI still lacks: heresy — _akpiper · 2026-11-11
- Harvard economist Deming: use AI to polish writing and you lose the learning — paulnovosad · 2026-11-11
- Tyler Cowen's conversation with Terence Tao is out, covering AI and beyond — tobias_rees · 2026-11-11
- Elvis Saravia: agents will subsume traditional software — and can even cut code bloat — omarsar0 · 2026-11-11