The 1% Rule for AI
What it is
The 1% Rule for AI is a personal-practice discipline that replaces the ambition to be wholesale-better at AI with the smallest unit of progress a leader will actually make. Three constraints carry the weight: one dimension small enough to name in a sentence and to test by Friday; embedded before the next, used three or four times in real work before another is added; one per week, no more. The compounding arithmetic (1.01 to the power of 52 working weeks is roughly 1.67) produces around a 67% capability gain by year-end against a leader who started in the same place and stayed there.
What it is
The 1% Rule for AI is a personal-practice discipline that replaces the ambition to be wholesale-better at AI with the smallest unit of progress a leader will actually make. Three constraints carry the weight: one dimension small enough to name in a sentence and to test by Friday; embedded before the next, used three or four times in real work before another is added; one per week, no more. The compounding arithmetic (1.01 to the power of 52 working weeks is roughly 1.67) produces around a 67% capability gain by year-end against a leader who started in the same place and stayed there.
Why it happens with AI
AI use is invisible by default, so without a tracked log the leader has no way to tell, on Friday, whether anything has changed. Three forces pull leaders away from the weekly micro-improvement: the size of the gap between current and aspired practice makes 1% feel trivially small; cultural pressure to perform AI mastery rewards visible action over invisible accumulation; and the absence of a tracked practice log dissolves the week into ‘a busy week with some AI in it’. The default outcome is a six-month plateau followed by indefinite repetition.
What working on it does, impact and benefits
Practising the rule replaces ambition with cadence. The leader who held the discipline for six months can name twenty-six things they are doing differently with AI; the leader who tried to be wholesale-better has read more and changed less. Over time the relationship inverts and the leader begins to teach AI the standards, vocabulary and judgement of their own work; the compounding accelerates when this second mode kicks in. The rule does not solve the strategic AI questions; it improves the asker.
Canonical framework: virenlall.com/one-percent-rule-ai, the full ~600-word treatment of the three constraints, the four-line Friday note, the three closing prompts, and the three embedded habits.
‘One named improvement a week, embedded before the next; the smallest unit of progress that will actually be made, held long enough to compound.’ — Viren Lall, Managing Director, ChangeSchool LDN (2026).