Reddit user questions AI job replacement after automating work with Claude
A Reddit user said they had automated part of their office job with Claude while questioning the claim that one AI-enabled worker could replace 20 employees. They described an Outlook outage that required calls, personal Gmail, WhatsApp, and Teams, and worried about workplace agents becoming unavailable if their underlying LLM failed. The account is an anecdotal community discussion rather than a verified deployment study.
The post offers a practitioner perspective on a practical constraint of workplace AI agents: dependence on external model and communication services can complicate continuity when systems are unavailable.
I have a typical office job where I speak to suppliers & customers from different countries, and various government agencies in UK. With Outlook being down I had to call people, tell them what my personal gmail address is, share information and documents via WhatsApp, etc. Luckily my Teams was working and I was able to call them, but it had been down - I would've been screwed. When my company replaced all the physical desk phones with Teams - I thought it was an idiotic idea, and I was right. And I say that as someone who's automated how of my job with Claude.
What happens when you have AI agents running the things and your LLM dies somewhere in Texas? I keep hearing this idea that a smart guy with AI can replace 20 employees.