← Full glossary
AI limbo
personasAI adoption limbo
AI limbo is the state in which an organization already uses or announces AI agents but has not trained its people, assigned owners, or communicated how work will change.
Term coined by People Managing People in its AI Limbo Report (December 2025 surveys of 1,000 workers and 379 HR professionals), adopted here as a descriptive label, not a standard. Report figures: 44% of workers say they received no agent training, 42% say their employer has not even mentioned agents, and 16% of HR professionals say nobody is responsible for oversight.
How it differs from
- Change management
- Change management is the discipline that prevents it; limbo is the state of not having done it.
- Resistance to change
- Resistance is a reaction by people; limbo is a gap in the organization.
Examples
- A company switches on a support agent without telling the team what it will do and what people keep.
- A pilot in production with no named owner or usage guide.
FAQ
- How do you get out of limbo?
- Name an owner for the agents, train the affected teams, and communicate what changes and what does not in each role.
- Is the data representative?
- It is a survey by an HR content publisher; it shows a trend, not a census.