When Intelligence Becomes Abundant

I recently had my OpenAI Codex usage quota increased by 10x, and it has meaningfully changed the way I work.

I can now use the highest-capability Codex modes across nearly everything (hello Codex 5.5 Extra-High /Fast): coding, user research, PRDs, strategy docs, planning, analysis, and the many small tasks that usually slow down a day.

Not having to ration intelligence is liberating.
When access to frontier models becomes effectively abundant, you start using them differently. You stop asking, “Is this worth the tokens?” and start asking, “What would I delegate if this capability were always available?”

Not having to ration intelligence is liberating.

This 10x increase gives me a small glimpse into what it must feel like inside frontier AI labs, where teams have broad access to the best models and can apply them deeply across their daily work.

That kind of access helps automate what is repetitive, compress what is tedious, and free up more time for the work that is strategic, creative, and meaningful.

Near-unlimited access to frontier models gives us a preview of how knowledge work itself is going to change.