The AI Control Center is the central place to make Enneo’s AI agent performance and control transparent. It shows how reliable AI agents are working, the degree of automation, and where there is further potential.Automation speeds up the handling of customer concerns, reduces process costs and relieves service staff who can focus on complex cases.The control center makes visible which business transactions are already being taken over by AI, where human approvals make sense and what steps towards full autonomous processing are possible.It makes this development measurable, comprehensible, and controllable, thus providing the key to gradually expanding automation in customer service.

Prerequisites for viewing and using

  • General Access: Permission in the Data and Reports → AI Control Center area
  • Automation settings: Permissions in the Dark Processing area control access for configuration, ticket detection, and triggering
  • Quality Inspection: Permission in the AI Quality Check area controls access to viewing, conducting test runs, and editing them
  • Live Overview: Permission in the Tickets → Ticket Overview Page area
  • AI Agent Team: Permission in the AI Agent Management area controls access to views, creation, and editing possibilities

Overview of Tools

The AI Control Center combines the most important automation tools in one place:
  • Live Overview & Metrics: Overview of live figures, degree of automation, and AI agent performance → learn more
  • Quality Check: Administration of test cases and evaluation of test runs → learn more
  • Automation Settings: Control of automation levels with and without release → learn more

Stages of Automation

The goal is fully autonomous processing of business transactions by AI agents. The journey towards this happens in several stages:
  1. Manual Processing
    The AI agent fully recognizes the issue, correctly reads relevant data, deduces suitable handling options, and provides them. A human decides on further action(s).
  2. Dark Processing with Approval
    The AI agent undertakes all steps of processing: It recognizes the issue, reads the data, takes the correct action, and selects the appropriate processing option. The verifier only needs to approve the choice made by the AI agent with a click, or reset the operation to manual processing if deviations are noticeable.
  3. Dark Processing without Approval
    The AI agent handles the processing entirely autonomously: It recognizes the issue, reads the data, deduces the correct action, selects the appropriate processing option and executes it independently.
The automation stages build upon each other.Only when the first stage works reliably can the transition to the next one make sense.Especially in the first two stages, it is crucial not merely to close cases, but to consciously analyze and evaluate them.In this way, it can be determined whether and at what points the AI agent should be further adjusted, or whether the next automation stage can be released.

Example: Bank Data Agent

A rule-based AI agent was created for the purpose of processing customer concerns about changes to bank details. It is supposed to check whether the IBAN provided by the customer is already stored in the system, whether it is invalid, or whether it can be adopted as a new, valid bank connection. Initially, an operator manually decides on the performance of the AI agent:
  • Is the customer’s concern correctly recognized (change of bank account details)?
  • Are all relevant parameters correctly read out (IBAN, validity date, account holder)?
  • Are the typical use cases correctly covered? (Valid IBAN, invalid IBAN, IBAN already exists, validity date in the past)
  • Are the correct processing options offered? (Adopt IBAN, inform customer about error in the IBAN, adjust validity date)
If the operator finds that individual points of the checklist are not reliably met, this is an indication that the Bank data agent needs to be adjusted. This includes adjustments to input parameter recognition, optimization of the business logic, or additions to the output handling. Only when these feedback items are implemented and the results are stable does the next step in automation happen. If the Bank data agent demonstrates consistently correct results, it can be set to Dark Processing with Approval. If reliable results continue to be confirmed here, the transition to Dark Processing without Approval, i.e., fully autonomous processing, follows.

Requirements and Risks

Before an AI agent is transferred to higher automation stages, certain conditions must be met:
  • Reliable customer recognition
  • Accurate concern recognition
  • Low error rates in recognizing and reading out parameters
  • Successful test runs in quality inspection
Risks arise mainly when incorrect decisions unnoticedly enter dark processing. The dashboard and quality inspection help minimize these risks and detect errors in time.

The Role of Users in the Control Centre

Users of the AI Control Centre have various tasks:
  • Create test cases and maintain to verify the reliability of the AI agents
  • Observe results and document and/or communicate any abnormalities
  • Make an assessment as to whether an AI agent can be transferred to the next automation stage.