---
id: video/eA7CENx7se0
type: video
title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
summary: "Copilot Test Toolkit in Business Central (2024 release wave 2), introduced in preview: an AL extension for testing Copilot features with data-driven tests, AI test context, Azure AI Studio external evaluation and multi-turn tests. Demonstrated on sales line suggestion and marketing text."
tier: official
language: en
tags:
  - copilot testing
  - large language models
  - data-driven testing
  - test toolkit
  - prompt engineering
  - responsible ai
  - test automation
  - sales line suggestion
  - ai test context
  - external evaluation
  - azure ai studio
  - multi-turn testing
system: copilot
review:
  state: reviewed
  by: opus
  at: "2026-10-07T23:03:25.870Z"
  flags: []
generated:
  at: "2026-10-07T23:03:25.916Z"
  pipeline: 0.2.0
  prompts:
    extract-video: 1
    summarize-video: 2
  input_hash: 37fe7268863bb4c9459987c2c302f5b3da90ad32a67b403fd4ed16efa201465f
evidence:
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=506s
    title: "Copilot Test Toolkit: preview"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 506
    quote: in this release, we are introducing the Copilot Test Toolkit in preview that allows you to test Copilot features
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=57s
    title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 57
    quote: there are some new considerations in the world of large language models and co-pilot features that require attention due to the high degree of
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=457s
    title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 457
    quote: large language models are sensitive to input variations and we want to cover how well the prompt generalizes and adapts to various contexts
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=506s
    title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 506
    quote: in this release, we are introducing the Copilot Test Toolkit in preview that allows you to test Copilot features in an efficient way
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=520s
    title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 520
    quote: It's an AL extension that can be installed in a sandbox environment, and you can find the source for this in our BC apps
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=794s
    title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 794
    quote: Each test procedure will be executed as many times as you have the entries in the data set.
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=940s
    title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 940
    quote: the question and the expected data, these are specific values supported by the tool, but you can format this however you want in your
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=1107s
    title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 1107
    quote: for this specific one, we are mocking the response, so there's no tokens consumed, but you would be able to see how many tokens
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=1155s
    title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 1155
    quote: our choice for external evaluation is Azure AI Studio. In Azure AI Studio, you can assess and compare AI application performance using built-in evaluators
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=1397s
    title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 1397
    quote: In multi-turn, you might have multiple inputs and outputs per test case. This is complex conversational scenarios.
  - kind: video
    url: https://www.youtube.com/watch?v=eA7CENx7se0&t=1481s
    title: "What's New: Testing Copilot in Business Central (2024 release wave 2)"
    date: "2024-10-09T13:47:41.000Z"
    commit: null
    t: 1481
    quote: Ensure that your data sets have a substantial number of entries. As mentioned before, even slight variability in prompts can impact how your feature
links:
  learn: []
  objects: []
  features: []
  topics: []
  localizations: []
  videos: []
  posts: []
  guidelines: []
video_id: eA7CENx7se0
channel: yt-microsoft
source_name: Microsoft Dynamics 365 Business Central (YouTube)
url: https://www.youtube.com/watch?v=eA7CENx7se0
published_at: "2024-10-09T13:47:41.000Z"
duration_s: 1588
captions: full
audience:
  - developer
  - partner
  - functional consultant
chapters:
  - t: 0
    title: Introduction to Copilot Testing
  - t: 26
    title: Challenges in Testing Large Language Models
  - t: 157
    title: Test Case Creation and Azure AI Studio
  - t: 290
    title: Types of Tests and Data-Driven Testing
  - t: 443
    title: Introducing the Copilot Test Toolkit
  - t: 533
    title: Copilot Test Toolkit Interface and Components
  - t: 670
    title: Test Suite Actions, Data Set Format, and Architecture
  - t: 851
    title: From Manual Tests to Data-Driven Testing with AI Test Context
  - t: 1046
    title: Running Tests and Evaluating Results in Business Central
  - t: 1144
    title: External Evaluation with Azure AI Studio
  - t: 1300
    title: Multi-Turn Testing for Conversational Scenarios
  - t: 1412
    title: Best Practices and Future Roadmap
features:
  - name: Copilot Test Toolkit
    status: preview
    t: 506
    verified: true
    status_source: video
  - name: Sales Line Suggestion Feature
    status: unclear
    t: 851
    verified: false
    status_source: video
  - name: Data-Driven Testing for Copilot
    status: unclear
    t: 415
    verified: false
    status_source: video
  - name: Accuracy Testing for Copilot
    status: unclear
    t: 303
    verified: false
    status_source: video
  - name: Quality Testing for Copilot
    status: unclear
    t: 356
    verified: false
    status_source: video
  - name: AI Test Context Code Unit
    status: unclear
    t: 968
    verified: false
    status_source: video
  - name: External Evaluation with Azure AI Studio
    status: unclear
    t: 1144
    verified: false
    status_source: video
  - name: Marketing Text Feature
    status: unclear
    t: 1200
    verified: false
    status_source: video
  - name: Multi-Turn Testing Support
    status: unclear
    t: 1374
    verified: false
    status_source: video
  - name: Test Suite Import and Export
    status: unclear
    t: 1519
    verified: false
    status_source: video
  - name: API Pages for Test Toolkit
    status: unclear
    t: 1519
    verified: false
    status_source: video
objects_mentioned:
  - codeunit AAT test context
  - codeunit AI test context
  - codeunit AI test
  - page AI test suites
quotes:
  - t: 57
    text: there are some new considerations in the world of large language models and co-pilot features that require attention due to the high degree of
    check: exact
  - t: 457
    text: large language models are sensitive to input variations and we want to cover how well the prompt generalizes and adapts to various contexts
    check: exact
  - t: 506
    text: in this release, we are introducing the Copilot Test Toolkit in preview that allows you to test Copilot features in an efficient way
    check: exact
  - t: 520
    text: It's an AL extension that can be installed in a sandbox environment, and you can find the source for this in our BC apps
    check: exact
  - t: 794
    text: Each test procedure will be executed as many times as you have the entries in the data set.
    check: exact
  - t: 940
    text: the question and the expected data, these are specific values supported by the tool, but you can format this however you want in your
    check: exact
  - t: 1107
    text: for this specific one, we are mocking the response, so there's no tokens consumed, but you would be able to see how many tokens
    check: exact
  - t: 1155
    text: our choice for external evaluation is Azure AI Studio. In Azure AI Studio, you can assess and compare AI application performance using built-in evaluators
    check: exact
  - t: 1397
    text: In multi-turn, you might have multiple inputs and outputs per test case. This is complex conversational scenarios.
    check: exact
  - t: 1481
    text: Ensure that your data sets have a substantial number of entries. As mentioned before, even slight variability in prompts can impact how your feature
    check: exact
---

# What's New: Testing Copilot in Business Central (2024 release wave 2)

> Copilot Test Toolkit in Business Central (2024 release wave 2), introduced in preview: an AL extension for testing Copilot features with data-driven tests, AI test context, Azure AI Studio external evaluation and multi-turn tests. Demonstrated on sales line suggestion and marketing text.

[Watch on YouTube](https://www.youtube.com/watch?v=eA7CENx7se0) · Microsoft Dynamics 365 Business Central (YouTube) · 2024-10-09 · 26:28 · tier official · reviewed (checked by Opus)

## Overview

The video explains why testing Copilot features differs from normal testing: large language models are sensitive to input variations, so the same test needs to run against many inputs. It introduces the Copilot Test Toolkit, an AL extension installed in a sandbox, with source in the BC apps repo. It is presented as being introduced in preview.

The session covers the toolkit interface, test suites, the data set format, the AI test context codeunit, and running tests in Business Central. It demos sales line suggestion (28 tests, 27 passing, 1 failing) and marketing text, which is evaluated externally in Azure AI Studio. It then covers multi-turn testing for conversations and best practices, and ends with a roadmap toward test automation.

## Key points

- The Copilot Test Toolkit is an AL extension, installed in a sandbox, with source in the BC apps repo (aka.ms/bcapps, AI Test Toolkit folder). It is introduced in preview in this release.
- Data-driven testing runs each test procedure once per entry in a data set, which covers LLM sensitivity to input variations.
- The AI test context codeunit gives test procedures their inputs from the data set (question, answer, context, ground truth, expected data) and can set test output for external evaluation.
- Data sets are uploaded in JSON lines format and can be marked as sensitive. Test suites can be imported and exported as XML files, and results can be downloaded for external evaluation.
- Marketing text is evaluated through Azure AI Studio using built-in or custom evaluators. This is subject to separate billing, and different metrics need different input sets.
- Multi-turn testing supports several inputs and outputs per test case for conversational scenarios. Calling the next turn API on the AI test context moves the test to the next question.
- The demos mock responses, so no tokens are consumed. Use data sets with many entries, because small prompt changes can change results.

## Chapters

- [0:00](https://www.youtube.com/watch?v=eA7CENx7se0&t=0s) Introduction to Copilot Testing
- [0:26](https://www.youtube.com/watch?v=eA7CENx7se0&t=26s) Challenges in Testing Large Language Models
- [2:37](https://www.youtube.com/watch?v=eA7CENx7se0&t=157s) Test Case Creation and Azure AI Studio
- [4:50](https://www.youtube.com/watch?v=eA7CENx7se0&t=290s) Types of Tests and Data-Driven Testing
- [7:23](https://www.youtube.com/watch?v=eA7CENx7se0&t=443s) Introducing the Copilot Test Toolkit
- [8:53](https://www.youtube.com/watch?v=eA7CENx7se0&t=533s) Copilot Test Toolkit Interface and Components
- [11:10](https://www.youtube.com/watch?v=eA7CENx7se0&t=670s) Test Suite Actions, Data Set Format, and Architecture
- [14:11](https://www.youtube.com/watch?v=eA7CENx7se0&t=851s) From Manual Tests to Data-Driven Testing with AI Test Context
- [17:26](https://www.youtube.com/watch?v=eA7CENx7se0&t=1046s) Running Tests and Evaluating Results in Business Central
- [19:04](https://www.youtube.com/watch?v=eA7CENx7se0&t=1144s) External Evaluation with Azure AI Studio
- [21:40](https://www.youtube.com/watch?v=eA7CENx7se0&t=1300s) Multi-Turn Testing for Conversational Scenarios
- [23:32](https://www.youtube.com/watch?v=eA7CENx7se0&t=1412s) Best Practices and Future Roadmap

## Features

| Feature | Status | At | Evidence |
|---|---|---|---|
| Copilot Test Toolkit | preview, demoed | [8:26](https://www.youtube.com/watch?v=eA7CENx7se0&t=506s) | "in this release, we are introducing the Copilot Test Toolkit in preview that allows you to test Copilot features" ([8:26](https://www.youtube.com/watch?v=eA7CENx7se0&t=506s)) |
| Sales Line Suggestion Feature | status not stated, demoed | [14:11](https://www.youtube.com/watch?v=eA7CENx7se0&t=851s) |  |
| Data-Driven Testing for Copilot | status not stated | [6:55](https://www.youtube.com/watch?v=eA7CENx7se0&t=415s) |  |
| Accuracy Testing for Copilot | status not stated | [5:03](https://www.youtube.com/watch?v=eA7CENx7se0&t=303s) |  |
| Quality Testing for Copilot | status not stated | [5:56](https://www.youtube.com/watch?v=eA7CENx7se0&t=356s) |  |
| AI Test Context Code Unit | status not stated, demoed | [16:08](https://www.youtube.com/watch?v=eA7CENx7se0&t=968s) |  |
| External Evaluation with Azure AI Studio | status not stated, demoed | [19:04](https://www.youtube.com/watch?v=eA7CENx7se0&t=1144s) |  |
| Marketing Text Feature | status not stated, demoed | [20:00](https://www.youtube.com/watch?v=eA7CENx7se0&t=1200s) |  |
| Multi-Turn Testing Support | status not stated, demoed | [22:54](https://www.youtube.com/watch?v=eA7CENx7se0&t=1374s) |  |
| Test Suite Import and Export | status not stated, demoed | [25:19](https://www.youtube.com/watch?v=eA7CENx7se0&t=1519s) |  |
| API Pages for Test Toolkit | status not stated | [25:19](https://www.youtube.com/watch?v=eA7CENx7se0&t=1519s) |  |

## AL objects mentioned

As heard in the captions. A name that matches one object page by exact type and name links to it; the others stay as named.

- codeunit "AAT test context" at [13:25](https://www.youtube.com/watch?v=eA7CENx7se0&t=805s)
- codeunit "AI test context" at [13:25](https://www.youtube.com/watch?v=eA7CENx7se0&t=805s)
- codeunit "AI test" at [15:51](https://www.youtube.com/watch?v=eA7CENx7se0&t=951s)
- page "AI test suites" at [17:48](https://www.youtube.com/watch?v=eA7CENx7se0&t=1068s)

Not found in BC28-30: codeunit "AAT test context", codeunit "AI test context", codeunit "AI test", page "AI test suites".

## Quotes

- [0:57](https://www.youtube.com/watch?v=eA7CENx7se0&t=57s) "there are some new considerations in the world of large language models and co-pilot features that require attention due to the high degree of"
- [7:37](https://www.youtube.com/watch?v=eA7CENx7se0&t=457s) "large language models are sensitive to input variations and we want to cover how well the prompt generalizes and adapts to various contexts"
- [8:26](https://www.youtube.com/watch?v=eA7CENx7se0&t=506s) "in this release, we are introducing the Copilot Test Toolkit in preview that allows you to test Copilot features in an efficient way"
- [8:40](https://www.youtube.com/watch?v=eA7CENx7se0&t=520s) "It's an AL extension that can be installed in a sandbox environment, and you can find the source for this in our BC apps"
- [13:14](https://www.youtube.com/watch?v=eA7CENx7se0&t=794s) "Each test procedure will be executed as many times as you have the entries in the data set."
- [15:40](https://www.youtube.com/watch?v=eA7CENx7se0&t=940s) "the question and the expected data, these are specific values supported by the tool, but you can format this however you want in your"
- [18:27](https://www.youtube.com/watch?v=eA7CENx7se0&t=1107s) "for this specific one, we are mocking the response, so there's no tokens consumed, but you would be able to see how many tokens"
- [19:15](https://www.youtube.com/watch?v=eA7CENx7se0&t=1155s) "our choice for external evaluation is Azure AI Studio. In Azure AI Studio, you can assess and compare AI application performance using built-in evaluators"
- [23:17](https://www.youtube.com/watch?v=eA7CENx7se0&t=1397s) "In multi-turn, you might have multiple inputs and outputs per test case. This is complex conversational scenarios."
- [24:41](https://www.youtube.com/watch?v=eA7CENx7se0&t=1481s) "Ensure that your data sets have a substantial number of entries. As mentioned before, even slight variability in prompts can impact how your feature"

## Disclaimers in the video

- [8:26](https://www.youtube.com/watch?v=eA7CENx7se0&t=506s) preview: we are introducing the Copilot Test Toolkit in preview
- [19:38](https://www.youtube.com/watch?v=eA7CENx7se0&t=1178s) subject-to-change: It's subject to separate billing, but you can learn more about that at aka.ms/azureaistudio.
- [25:42](https://www.youtube.com/watch?v=eA7CENx7se0&t=1542s) coming-later: We aim to improve the tool to enable the test automation and make it even smoother for you to test your Copilot features with less effort.

Presenters (as heard): Peter Borring, Julia Møller, Steen Klitgård, Julian, Peter.