Asana's browser-agent cost drop shows where enterprise AI quietly compounds

Asana cutting browser-agent model costs by 76x with GPT-6.1 Sol is a good example of where enterprise AI can compound quietly: not in a flashy demo, but in repeated workflow tests getting cheaper.

Browser agents are expensive when every run consumes long context, retries, and visual state. Cost reductions change how often companies can test, evaluate, and deploy them across ordinary work.

The important question is whether lower costs preserve reliability. An inexpensive browser agent is valuable only if it completes tasks without making people babysit each click.