I have watched a pattern repeat itself often enough across organizations that I no longer think of it as an anecdote. I think of it as a test. It is not a test anyone administers on purpose. It reveals itself in what a leader does in the six months after an AI system removes a genuine burden from her team.
The burden is real. Forecasting cycles that took an analyst three days now take an afternoon. Reporting packages that consumed a Friday now assemble themselves by Wednesday. The leader who deployed the system did nothing wrong. The tool worked exactly as intended.
What happens next is the test.
In the pattern I have watched most often, the freed time gets absorbed the same way water finds a drain. The organization simply produces more reports, more forecasts, more of the same work at higher frequency. Nobody decided this. It happened because nobody decided anything else. The analyst who used to spend three days building a forecast now spends the same three days building three forecasts, none of them more useful than the one before, because the additional capacity was never pointed anywhere.
This is not a failure of the technology. The technology performed. It is a failure of leadership imagination, and it is the more common outcome by a wide margin.
The rarer pattern is the one worth naming. A leader I observed, running a finance function inside a mid-sized enterprise, watched the same AI system compress her team's forecasting cycle the same way. She did not ask her analysts to produce more forecasts. She asked a different question: which of these people is ready to run a business unit, and what is stopping me from finding out.
She used the recovered time to do something the forecasting cycle had never left room for. She sat with each analyst individually and asked what they thought the numbers meant, not just what the numbers said. She asked which decisions they would make with the information if the decision were theirs to make. Two of them had answers ready. They had been thinking about it for years and had never been asked.
Within eighteen months, both were running functions they had never been considered for under the old cadence, because the old cadence never produced a moment where anyone found out what they were capable of. The forecasting still happens faster than it used to. That was never the achievement. The achievement was a leader who understood that automation frees capacity, and capacity is only valuable if someone decides what it is for.
This is the Expectation Elevation Model in its plainest form. The highest function of a leader is not to manage the performance of the work in front of her. It is to use every advantage available, including the time an efficient system hands back, to help the people around her see more in themselves than they had seen before. A leader who only redeploys freed capacity toward more of the same output has automated a process. A leader who redeploys it toward the people has done something a system cannot do at all.
The distinction matters more now than it has in the past because AI systems are, for the first time, freeing meaningful blocks of senior time at scale. Every enterprise deploying agentic tools this year is running this same experiment, whether its leadership recognizes it or not. Most will fill the recovered hours with more activity. A few will fill them with the conversations that used to get postponed because there was never time. Those few will discover things about their people that the metrics were never built to show.
The Legacy Test applies here with unusual precision. It is not whether the forecasting cycle got faster. Every competitor's forecasting cycle is getting faster too, and none of that survives the leader who built it. What survives is the analyst who is now running a business unit because someone used freed time to find out what she was capable of, and who will go on to ask the same question of someone else, on her own initiative, without being told to. That is not a system. That is a standard, carried forward by a person, into rooms the original leader will never enter.
The full research on building the leadership architecture that makes this the default outcome rather than the rare one is developed in the Leadership Reinvention in the AI Era Executive Playbook.
Glenn E. Daniels II | Touch Stone Publishers