There is no AI FOMO
If you’re subscribed to the X echo chamber around all things tech, you couldn’t have possibly missed updates about Clawdbot/Moltbot and how you’re missing out if you’re not shelling out money to get a MacMini to run all these digital agents to do your bidding!
YOU ARE MISSING OUT ON NOTHING.
AI is both over-hyped and under-hyped right now. Thankfully, the real life changing aspects of AI will have an extremely long tail timeframe where you can play with it, learn it and use if effectively. In the span of a week, ClawdBot moved from a $900 Mac Mini investment to a $5/month Cloudflare worker investment – and likewise, economics of scale will continue to work in our favour for each new AI advancement that continues to deliver effective results (while most other experiments fail and wither away over time).
Until then, don’t jump from model to model, or from trend to trend in this space. I feel like breadth of knowledge will matter little in the long term. What I would do instead (and what I am doing) is picking any model/tool and going deep with it to build something. The ability to leverage AI tooling to make real world progress towards a goal is more impressive than knowing the subtle differences between the latest releases of Codex and Claude Code. If you’re looking for an old school metaphor, it matters less that you know if Ubuntu was better than Alpine in 2009 – what matters more if that you chose one distro and explored it inside out.