When my chainsaw gets too hot, I give it a rest and go off for a bit, then let it cool down before doing more work.
After all, I don’t want the motor to overheat and then have to buy a whole new one and have it cost even more.
Does the same apply to our AI interactions? Let AI go to town for a bit, but then put it on pause over lunch so it can cool down and come back at it?
Or do we keep expanding infinitum so it is always online, all the time, consuming everything?
Will our cost of AI become like electricity, where at certain parts of the day (early morning) it is much cheaper to use as opposed to during peak hours?
We have (thankfully) managed the iteration of consuming as many tokens as you can to demonstrate productivity – so what’s next for AI consumption patterns?
What’s the next step in giving AI Rest between our workloads?