--- id: video/juUBVPfIp8A type: video title: "What's New: AI Development Toolkit (2026 release wave 1)" summary: "AI Development Toolkit updates in Business Central 2026 release wave 1: the renamed toolkit, simpler BC AI Resources authentication, content filtering and prompt injection protection for Azure Open AI, and evaluation features (suite configurations, multi-language, Copilot credit insights, agent evaluation)." tier: official language: en tags: - ai development toolkit - bc ai resources - agents - azure open ai - content filtering - prompt injection - xpi detection - evaluations - multi-language support - copilot credits - agent evaluation - test context system: copilot review: state: reviewed by: opus at: "2026-10-07T22:49:50.869Z" flags: [] generated: at: "2026-10-07T22:49:50.911Z" pipeline: 0.2.0 prompts: extract-video: 1 summarize-video: 2 input_hash: 6c88e65a6944fed0490b135e4d0f49a9f5bbb13c5194d149a0f77b7377f6dd75 evidence: - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=55s title: "Agent development and design: preview" date: "2026-04-01T12:00:51.000Z" commit: null t: 55 quote: The latter has actually been in public preview since the beginning of February. - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=947s title: "Agent library: announced" date: "2026-04-01T12:00:51.000Z" commit: null t: 947 quote: very soon we'll have an agent library coming giving you even more capabilities to test agents - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=24s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 24 quote: the toolkit now does not just support the ability to use azure open ai to use llms from al code - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=55s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 55 quote: The latter has actually been in public preview since the beginning of February. We've already seen almost 1,500 agents being built by the community - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=209s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 209 quote: we have now removed that requirement. So, when you authenticate or authorize in code using the set managed resource authorization we now have a - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=338s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 338 quote: We have added support for only low and medium content filtering policies. Where low is the default and actually the the setup that will - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=371s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 371 quote: We do not support the high setting, basically passing the most harmful content. That is something that you can have if you're using your - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=576s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 576 quote: the first thing you'll probably notice is the terminology change from evaluation or from tests to evaluations. And this is also represented throughout the - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=707s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 707 quote: you can set the run frequency to daily or weekly or manual. And this is an extensible enum, so you can extend with your - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=800s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 800 quote: you can now see that they are categorized by languages - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=874s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 874 quote: Copilot credit insights to help you understand AI consumptions for evaluations - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=922s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 922 quote: This is still in the early stages, but now you can build and validate AI agents in BC - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=947s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 947 quote: very soon we'll have an agent library coming giving you even more capabilities to test agents - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=1059s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 1059 quote: we now have changed the name of what was before the development tools for extending Copilot Business Central or just a Copilot tool kit - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=1074s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 1074 quote: you no longer have to pass the your partner as your OpenAI subscription when you are using the built-in AI resources - kind: video url: https://www.youtube.com/watch?v=juUBVPfIp8A&t=1088s title: "What's New: AI Development Toolkit (2026 release wave 1)" date: "2026-04-01T12:00:51.000Z" commit: null t: 1088 quote: flexibility you have for setting the Azure OpenAI policies specifically for content filtering on harmful content as well as um the X P I links: learn: [] objects: [] features: [] topics: [] localizations: [] videos: [] posts: [] guidelines: [] video_id: juUBVPfIp8A channel: yt-microsoft source_name: Microsoft Dynamics 365 Business Central (YouTube) url: https://www.youtube.com/watch?v=juUBVPfIp8A published_at: "2026-04-01T12:00:51.000Z" duration_s: 1206 captions: full audience: - developer - partner - functional consultant chapters: - t: 0 title: AI Development Toolkit Overview - t: 85 title: Learning Resources - t: 125 title: BC AI Resources and Authentication - t: 269 title: Content Filtering Policies - t: 435 title: Prompt Injection Protection - t: 546 title: AI Evaluations Overview - t: 656 title: Evaluation Suite Configurations - t: 787 title: Multi-Language Support and Copilot Credit Insights - t: 922 title: Agent Evaluation Support - t: 1021 title: Summary of AI Development Toolkit updates - t: 1130 title: Resources and Community Engagement features: - name: AI Development Toolkit renamed status: unclear t: 6 verified: false status_source: video - name: Agent development and design status: preview t: 44 verified: true status_source: video - name: BC AI Resources authentication simplified status: unclear t: 168 verified: false status_source: video - name: Content filtering policies status: unclear t: 269 verified: false status_source: video - name: Azure Open AI policy parameters codeunit status: unclear t: 412 verified: false status_source: video - name: Prompt injection protection status: unclear t: 435 verified: false status_source: video - name: AI evaluation tool renamed to evaluation status: unclear t: 546 verified: false status_source: video - name: Evaluation suite configurations status: unclear t: 656 verified: false status_source: video - name: Multi-language evaluation support status: unclear t: 753 verified: false status_source: video - name: Copilot credit insights in evaluations status: unclear t: 864 verified: false status_source: video - name: Agent evaluation support status: unclear t: 922 verified: false status_source: video - name: Agent library status: announced t: 947 verified: true status_source: video objects_mentioned: - codeunit Azure Open AI policy parameters - codeunit agent test test context code unit - other test context quotes: - t: 24 text: the toolkit now does not just support the ability to use azure open ai to use llms from al code check: fuzzy - t: 55 text: The latter has actually been in public preview since the beginning of February. We've already seen almost 1,500 agents being built by the community check: exact - t: 209 text: we have now removed that requirement. So, when you authenticate or authorize in code using the set managed resource authorization we now have a check: exact - t: 338 text: We have added support for only low and medium content filtering policies. Where low is the default and actually the the setup that will check: exact - t: 371 text: We do not support the high setting, basically passing the most harmful content. That is something that you can have if you're using your check: exact - t: 576 text: the first thing you'll probably notice is the terminology change from evaluation or from tests to evaluations. And this is also represented throughout the check: exact - t: 707 text: you can set the run frequency to daily or weekly or manual. And this is an extensible enum, so you can extend with your check: exact - t: 800 text: you can now see that they are categorized by languages check: exact - t: 874 text: Copilot credit insights to help you understand AI consumptions for evaluations check: exact - t: 922 text: This is still in the early stages, but now you can build and validate AI agents in BC check: exact - t: 947 text: very soon we'll have an agent library coming giving you even more capabilities to test agents check: exact - t: 1059 text: we now have changed the name of what was before the development tools for extending Copilot Business Central or just a Copilot tool kit check: exact - t: 1074 text: you no longer have to pass the your partner as your OpenAI subscription when you are using the built-in AI resources check: exact - t: 1088 text: flexibility you have for setting the Azure OpenAI policies specifically for content filtering on harmful content as well as um the X P I check: exact --- # What's New: AI Development Toolkit (2026 release wave 1) > AI Development Toolkit updates in Business Central 2026 release wave 1: the renamed toolkit, simpler BC AI Resources authentication, content filtering and prompt injection protection for Azure Open AI, and evaluation features (suite configurations, multi-language, Copilot credit insights, agent evaluation). [Watch on YouTube](https://www.youtube.com/watch?v=juUBVPfIp8A) · Microsoft Dynamics 365 Business Central (YouTube) · 2026-04-01 · 20:06 · tier official · reviewed (checked by Opus) ## Overview The toolkit formerly known as the development tools for extending Copilot is now called the AI Development Toolkit. It covers using Azure Open AI from AL code and also designing and coding agents in Business Central. Agent development has been in public preview since February, with almost 1,500 agents built by the community. The video walks through authentication changes for BC AI Resources, content filtering policies, prompt injection protection, and the renamed evaluation tool. Evaluations gain suite configurations, language categorization of test inputs, Copilot credit insights, and early agent evaluation support. Most of these were demoed. ## Key points - The toolkit was renamed AI Development Toolkit; the AI test tool is now called AI Development Toolkit evaluation. - BC AI Resources no longer require your own partner subscription. The set managed resource authorization method has a new signature without subscription details; the old one is obsoleted but still works. - Content filtering supports low and medium levels, with low as the default. High is not supported in managed resources and needs your own Azure Open AI subscription with approval. Levels are set through the Azure Open AI policy parameters codeunit. - Prompt injection protection uses Azure AI content prompt shields. You must call the detection tags for untrusted prompts and enable XPI detection separately. - Evaluation suite configurations define capability, test type and frequency (daily, weekly, manual). The frequency enum is extensible, and pipelines must adopt the configurations manually because they do not run evaluations automatically. - Multi-language evaluations require name and language in data sets, and data set names must be common across languages. Copilot credit use shows per run, line and data set entry. - Agent evaluation is in the early stages and lets you inspect agent tasks per run, line and entry. An agent library is announced as coming soon. ## Chapters - [0:00](https://www.youtube.com/watch?v=juUBVPfIp8A&t=0s) AI Development Toolkit Overview - [1:25](https://www.youtube.com/watch?v=juUBVPfIp8A&t=85s) Learning Resources - [2:05](https://www.youtube.com/watch?v=juUBVPfIp8A&t=125s) BC AI Resources and Authentication - [4:29](https://www.youtube.com/watch?v=juUBVPfIp8A&t=269s) Content Filtering Policies - [7:15](https://www.youtube.com/watch?v=juUBVPfIp8A&t=435s) Prompt Injection Protection - [9:06](https://www.youtube.com/watch?v=juUBVPfIp8A&t=546s) AI Evaluations Overview - [10:56](https://www.youtube.com/watch?v=juUBVPfIp8A&t=656s) Evaluation Suite Configurations - [13:07](https://www.youtube.com/watch?v=juUBVPfIp8A&t=787s) Multi-Language Support and Copilot Credit Insights - [15:22](https://www.youtube.com/watch?v=juUBVPfIp8A&t=922s) Agent Evaluation Support - [17:01](https://www.youtube.com/watch?v=juUBVPfIp8A&t=1021s) Summary of AI Development Toolkit updates - [18:50](https://www.youtube.com/watch?v=juUBVPfIp8A&t=1130s) Resources and Community Engagement ## Features | Feature | Status | At | Evidence | |---|---|---|---| | AI Development Toolkit renamed | status not stated | [0:06](https://www.youtube.com/watch?v=juUBVPfIp8A&t=6s) | | | Agent development and design | preview | [0:44](https://www.youtube.com/watch?v=juUBVPfIp8A&t=44s) | "The latter has actually been in public preview since the beginning of February." ([0:55](https://www.youtube.com/watch?v=juUBVPfIp8A&t=55s)) | | BC AI Resources authentication simplified | status not stated | [2:48](https://www.youtube.com/watch?v=juUBVPfIp8A&t=168s) | | | Content filtering policies | status not stated, demoed | [4:29](https://www.youtube.com/watch?v=juUBVPfIp8A&t=269s) | | | Azure Open AI policy parameters codeunit | status not stated, demoed | [6:52](https://www.youtube.com/watch?v=juUBVPfIp8A&t=412s) | | | Prompt injection protection | status not stated, demoed | [7:15](https://www.youtube.com/watch?v=juUBVPfIp8A&t=435s) | | | AI evaluation tool renamed to evaluation | status not stated | [9:06](https://www.youtube.com/watch?v=juUBVPfIp8A&t=546s) | | | Evaluation suite configurations | status not stated, demoed | [10:56](https://www.youtube.com/watch?v=juUBVPfIp8A&t=656s) | | | Multi-language evaluation support | status not stated, demoed | [12:33](https://www.youtube.com/watch?v=juUBVPfIp8A&t=753s) | | | Copilot credit insights in evaluations | status not stated, demoed | [14:24](https://www.youtube.com/watch?v=juUBVPfIp8A&t=864s) | | | Agent evaluation support | status not stated, demoed | [15:22](https://www.youtube.com/watch?v=juUBVPfIp8A&t=922s) | | | Agent library | announced | [15:47](https://www.youtube.com/watch?v=juUBVPfIp8A&t=947s) | "very soon we'll have an agent library coming giving you even more capabilities to test agents" ([15:47](https://www.youtube.com/watch?v=juUBVPfIp8A&t=947s)) | ## 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 "Azure Open AI policy parameters" at [7:03](https://www.youtube.com/watch?v=juUBVPfIp8A&t=423s) - codeunit "agent test test context code unit" at [16:37](https://www.youtube.com/watch?v=juUBVPfIp8A&t=997s) - other "test context" at [16:49](https://www.youtube.com/watch?v=juUBVPfIp8A&t=1009s) Not found in BC28-30: codeunit "Azure Open AI policy parameters", codeunit "agent test test context code unit". ## Quotes - [0:24](https://www.youtube.com/watch?v=juUBVPfIp8A&t=24s) "the toolkit now does not just support the ability to use azure open ai to use llms from al code" - [0:55](https://www.youtube.com/watch?v=juUBVPfIp8A&t=55s) "The latter has actually been in public preview since the beginning of February. We've already seen almost 1,500 agents being built by the community" - [3:29](https://www.youtube.com/watch?v=juUBVPfIp8A&t=209s) "we have now removed that requirement. So, when you authenticate or authorize in code using the set managed resource authorization we now have a" - [5:38](https://www.youtube.com/watch?v=juUBVPfIp8A&t=338s) "We have added support for only low and medium content filtering policies. Where low is the default and actually the the setup that will" - [6:11](https://www.youtube.com/watch?v=juUBVPfIp8A&t=371s) "We do not support the high setting, basically passing the most harmful content. That is something that you can have if you're using your" - [9:36](https://www.youtube.com/watch?v=juUBVPfIp8A&t=576s) "the first thing you'll probably notice is the terminology change from evaluation or from tests to evaluations. And this is also represented throughout the" - [11:47](https://www.youtube.com/watch?v=juUBVPfIp8A&t=707s) "you can set the run frequency to daily or weekly or manual. And this is an extensible enum, so you can extend with your" - [13:20](https://www.youtube.com/watch?v=juUBVPfIp8A&t=800s) "you can now see that they are categorized by languages" - [14:34](https://www.youtube.com/watch?v=juUBVPfIp8A&t=874s) "Copilot credit insights to help you understand AI consumptions for evaluations" - [15:22](https://www.youtube.com/watch?v=juUBVPfIp8A&t=922s) "This is still in the early stages, but now you can build and validate AI agents in BC" - [15:47](https://www.youtube.com/watch?v=juUBVPfIp8A&t=947s) "very soon we'll have an agent library coming giving you even more capabilities to test agents" - [17:39](https://www.youtube.com/watch?v=juUBVPfIp8A&t=1059s) "we now have changed the name of what was before the development tools for extending Copilot Business Central or just a Copilot tool kit" - [17:54](https://www.youtube.com/watch?v=juUBVPfIp8A&t=1074s) "you no longer have to pass the your partner as your OpenAI subscription when you are using the built-in AI resources" - [18:08](https://www.youtube.com/watch?v=juUBVPfIp8A&t=1088s) "flexibility you have for setting the Azure OpenAI policies specifically for content filtering on harmful content as well as um the X P I" ## Disclaimers in the video - [0:55](https://www.youtube.com/watch?v=juUBVPfIp8A&t=55s) preview: The agent development capability has been in public preview since the beginning of February. - [15:22](https://www.youtube.com/watch?v=juUBVPfIp8A&t=922s) preview: agent evaluation support is still in the early stages - [15:47](https://www.youtube.com/watch?v=juUBVPfIp8A&t=947s) coming-later: agent library coming giving you even more capabilities to test agents - [17:12](https://www.youtube.com/watch?v=juUBVPfIp8A&t=1032s) preview: agents that we just introduced in public preview Presenters (as heard): Peter, Steen, Steve.