"Everybody will have to use AI. Because if you don't use AI, you will lose your job to somebody that does" -- Jenson Huang, CEO Nvidia I keep hearing this exact tension in conversations with friends, family, and coworkers. The moment AI comes up, the vibe shifts: “It’s going to take all the jobs. It’s evil. We’re all doomed.” But what if we’re looking at it backwards? AI already holds within it essentially all the books humanity has ever written. It’s a tool that gives us access to knowledge far beyond any single human lifetime — compressing centuries of wisdom, science, creativity, and experience into seconds. Instead of replacing us, it can expand us. It lets us ask better questions, solve harder problems, and explore realms we couldn’t reach alone. Every major technology in history (fire, electricity, computers, the internet) was feared for the jobs it would destroy. Yet each one ultimately created far more opportunity than it eliminated — for those who learned to wield it. The key isn’t resisting AI. It’s learning to guide it intelligently. Just like effective prompting: you don’t tell the model *how* to do something step-by-step. You clearly describe the outcome you want. The same principle applies to our relationship with AI as a society. We decide the direction. We set the guardrails. We choose whether it amplifies human potential or something else. --------------------------------- Edited: This just dropped from AWS CEO The “AI will take all jobs” story is too simple. Some jobs will shrink. Some will change. Some new ones will appear. But the real divide may be behavioral. People who avoid AI will feel hunted by it. People who test it early will see where it helps. That does not make the transition painless. It just makes panic a bad strategy. --------------------------------- So I’m curious — where do you stand? - Are you already using AI regularly in your work or life? - Or do you still see it mainly as a threat? - What’s the best way you’ve found to reframe this conversation with people who are skeptical or afraid?

Curious what people think. I’ve maxed out my current Pro usage, so next week I’m considering upgrading to Cursor Ultra, but I’m also wondering whether Claude Max would be the better move. The reason I’m thinking about this is that the system I’m building is adapter-based, so in theory I can swap between providers/models depending on availability. My question is: if Claude hits its usage cap, could I simply switch the system over to a Cursor/OpenAI-style adapter and keep working from there, assuming the adapters are set up properly? Basically trying to understand whether the smarter setup is: - one higher-tier tool, - multiple subscriptions, - or a provider-adapter setup where I can route around usage limits. Would love to hear how others are handling caps, cost, and model switching in real build workflows.