WHERE DO HUMANS FIT IN AN AGENT?

The goal of an Agent is not to remove humans from every process.

Humans still need to decide:

• the goal;
• the method;
• which data is trustworthy;
• the Agent’s limits;
• the permissions it is allowed to use;
• which actions can be automated;
• which results need checking.

You can picture it as:

Humans design the method. The Agent carries out the method. Humans control the result.

Only once the level of trust has been proven should part of the workflow be automated more deeply.