A simple model for organising AI agents in your company: ownership, a general-purpose baseline and custom agents.
The short answer
Organise AI in three layers: decide internally who owns which agents, give everyone a general-purpose assistant for basics, and build custom agents for the workflows with business impact.
Key takeaways
Each agent needs a business owner responsible for its output, instructions and results.
A general-purpose assistant such as Claude or ChatGPT covers writing, research and summaries for everyone.
Demanding workflows get purpose-built agents. This model comes straight from how Switch frames agent organisations.
The leader of the team whose workflow it runs, not just IT.
Helpful at scale, but not required to start.
How this guide was made
Written and reviewed by our editors against vendor documentation, public regulation texts and practitioner interviews. We update guides when products or rules change. Tools mentioned may have a commercial relationship with this site; this never decides what we recommend.
Keep reading