We’re entering dangerous territory with AI
TL;DR
Vox journalist and AI reporter Kelsey Piper joins a deep-dive conversation on the current state of AI, separating real progress from hype. AI has evolved beyond answering questions to autonomously writing code — capabilities once exclusive to humans. Piper, now at The Argument, is an optimist about technology but stresses the serious risks of agentic AI systems that act independently. The discussion warns that society may be caught in familiar tech hype cycles, this time with stakes far higher than before.
Nauti's Take
Kelsey Piper is one of the sharpest AI thinkers out there, and her warning deserves attention: we might be in another hype cycle, but this time the stakes are orders of magnitude higher. Agentic AI isn't a future scenario anymore — it's happening now.
The uncomfortable question isn't whether something will go wrong, but whether anyone is actually in a position to stop it.
Context
Agentic AI represents a qualitative leap: when models stop advising and start acting, the questions of accountability, control, and societal risk shift fundamentally. The question is no longer 'Can AI do this? ' but 'Who is responsible when AI does this?
' Kelsey Piper's perspective is valuable precisely because it avoids both breathless boosterism and reflexive panic – instead offering a structured lens to cut through the hype cycle.
Sources