AI Practices Remain Trapped in Individual Documents at Organizations

AI Usage Patterns Concentrated Among Non-Technical Roles

At one financial infrastructure company, the top 10 token users were almost entirely non-technical staff including marketers, operations folks, and customer support. These individuals generated the most AI knowledge through daily work. Useful AI practices often get trapped in personal docs and Slack threads.

Roundtable Examples Show Isolated Prompt Development

At a recent roundtable, one participant described a nine-page Google Doc prompt for developing creative briefs that she pasted at the start of every session. She was the only one using this document at her company. A few minutes later, another participant noted the same pattern had occurred in three separate business lines with different marketing managers, resulting in no common language, no common skill, and no common brain.

Enablement Flows Primarily One Direction

Most AI enablement programs push knowledge outward from a central team that selects tools, develops training, and publishes playbooks. When the last time an AI program brought what employees were learning back to the people running it remains unaddressed, front-line discoveries stay with individuals. According to MarTech, this means organizations pay for the same lesson repeatedly.

Requirements for Bi-Directional Knowledge Spread

According to MarTech, turning scattered AI wins into organizational advantage requires an AI Lead whose job includes noticing signals such as a nine-page prompt mentioned in passing. It also needs a named function with authority to refine discoveries into usable formats and a horizontal path so one group’s solution reaches others facing the same issue. The guidewords listed are Spark, spread, scale, sustain.

A version of the nine-page Google Doc exists somewhere in many organizations, according to MarTech, built by someone who uses it daily yet unknown to the wider group.