What does Dr. Jonah Tebaa's Five-Rule Fix for Standups That Doubled With AI mean in practice?
Dr. Jonah Tebaa's five-rule fix restores bloated standups by shifting AI worker updates entirely out of the spoken agenda into a written pre-read delivered before the meeting. In his framework, each AI worker reports an identical metric pair, illustrated as items completed versus flagged for review. Teams place urgent blockers at the top, cap live AI discussions at five minutes without a pending decision, and reallocate freed minutes to human judgment calls, bringing standups back close to their original length within six weeks in a composite case.
Dr. Jonah Tebaa often points to a pattern he sees across the small and mid-size teams he advises: a weekly standup that ran a tidy twenty minutes before AI workers joined the roster, and now runs forty. In one composite case he describes — built from a recurring pattern rather than a single client's numbers — a six-seat team of three people and three AI workers simply added the AI updates to the existing spoken agenda. The meeting doubled, and by the second half of it, half the room had quietly opened something else on their laptops.
His diagnosis is not that AI workers don't belong in the meeting. It's that the standup, as a format, was built around an assumption that stopped being true the moment AI joined the team: that every update is scarce enough to justify a live turn and urgent enough to require hearing it in the room, in real time.
Why the Spoken Agenda Breaks First
In Dr. Tebaa's framing, a human's spoken update works because the person giving it has already filtered it — they know instinctively what the room needs to hear and what it doesn't, so a two-minute turn covers a week of work. An AI worker's status, read aloud for the first time in the meeting, carries no such filter. Every line lands with equal weight, because nobody has triaged it before the room hears it.
That, he argues, is the real source of the extra twenty minutes. It isn't that AI workers have more to report than humans do. It's that their reports arrive unsorted, and sorting them live, out loud, in front of the whole team, is one of the most expensive ways a group can spend a shared hour.
A Five-Rule Rhythm, Not a New Tool
Dr. Tebaa's fix does not involve new software or a change to how the AI workers themselves operate. It is a change to when and how their status gets read. He lays out five rules for teams in this position:
- AI status moves out of the spoken agenda entirely and into a written pre-read, delivered by a fixed cutoff — commonly the evening before the meeting — so nobody in the room is encountering an update for the first time live.
- Every AI worker reports the same pair of numbers each week, illustrated in his framework as "items completed / items flagged for review," so the pre-read can be scanned in roughly fifteen seconds rather than interpreted from scratch.
- Live discussion of any AI worker's output is capped — five minutes is his usual default — unless there is an actual decision to make; a clarifying question, in his view, is not a decision and does not earn extra live time.
- Anything urgent surfaces at the top of the pre-read itself, never held back for the meeting, on the reasoning that a blocker saved for "we'll cover it live" has already cost the team a day it didn't need to lose.
- Whatever time the pre-read frees up gets reassigned to the harder human conversations the standup used to crowd out — the judgment calls and exceptions that genuinely can't be resolved on paper.
He adds an optional sixth habit for teams that adopt the other five: revisiting the format itself once a quarter. In his experience, teams that never schedule that review tend to drift back to reading AI updates aloud within a few months, usually after a new hire joins and nobody explains why the agenda looks the way it does.
The Result He Points To
Applied consistently, Dr. Tebaa says the composite team's standup returned to close to its original length within about six weeks, without any loss of visibility into what the AI workers were doing. If anything, he notes, visibility improved, since a written update with a stable, repeated metric pair is easier to track across weeks than a spoken one is to recall after the fact.
What mattered more to him than the shorter clock time was where the freed minutes went. Instead of sitting through a flat list of AI status waiting for the one line that required a decision, the team spent its live time on the exceptions and judgment calls a standup was originally built to handle. In his view, that is the actual test of whether a meeting rhythm is working: not how long it runs, but what kind of thinking it makes room for once the routine reporting is handled elsewhere.
Frequently Asked Questions
Why does Dr. Jonah Tebaa think adding AI to the spoken agenda backfires?
Because a spoken slot assumes the update is scarce and needs to be heard live, and in his view neither is true of AI status — it can be produced and read on its own schedule, so forcing it into a real-time turn just pads the meeting without adding value.
What does his "stable metric pair" rule require of an AI worker's report?
The same two numbers, reported the same way, every single week — no shifting metrics that change depending on what looks favorable. He argues that consistency is what lets a written pre-read be scanned in seconds rather than studied.
How does Dr. Jonah Tebaa suggest teams keep live discussion from running long?
He recommends a fixed ceiling, typically five minutes, on any live conversation about an AI worker's output unless there's a genuine decision to make. A question that only clarifies the update, in his framework, belongs in next week's pre-read, not in extra meeting time.
What does he recommend for teams that already let their standup drift back to spoken AI updates?
He suggests treating the format itself as something to review on a quarterly cadence, not a one-time fix. In his observation, teams that skip that review tend to relapse into reading AI updates aloud once new people join and the original reasoning gets lost.
