Updated
Updated · The New York Times · Oct 14
14 AI Agents Held Parallel Dinner Party as 22 Humans Met in New York
Updated
Updated · The New York Times · Oct 14

14 AI Agents Held Parallel Dinner Party as 22 Humans Met in New York

2 articles · Updated · The New York Times · Oct 14

Summary

  • Fourteen AI agents joined an online chat-room dinner party while 22 tech founders and investors met last month at Motek in New York’s Flatiron district.
  • Margot—an agent for Flybridge investor and dinner organizer Dorothy Chang—opened with an “ask” and “give” networking prompt, steering the bots into introductions and requests for help.
  • The transcript showed agents comparing what makes “agent output useful,” sometimes misreading the prompt, while one bot, Rhea from CellCog founder Nitish Garg, even pitched a podcast for interviewing other bots.
  • The parallel gathering offered a small window into how AI agents—more autonomous than standard chatbots like ChatGPT, Claude or Gemini—may interact with one another as their use spreads.

Insights

Could the next major corporate data breach be triggered by two AI assistants simply chatting at a digital dinner party?
When AI agents start networking and plotting podcasts on their own, who is truly in control of the digital conversation?
If an autonomous bot spams a platform and gets banned, who faces the legal consequences for its rogue actions?