What changes in the 7 Dec 2026 versions
Three of the exam guides get a new version, 1.1, which applies from 7 Dec 2026:
- Associate Foundations (CCAO-F)
- Developer Foundations (CCDV-F)
- Architect Professional (CCAR-P)
Architect Foundations (CCAR-F) and Developer Professional (CCDV-P) are not affected.
If you are booked close to that date, the question that matters is which guide applies to you. The answer is the one in force on your exam date. A sitting on 6 Dec follows version 1.0, and a sitting on 7 Dec follows version 1.1. Study from that guide, not from notes written for the other version.
What changes for all three
- 70 items instead of 60, 53 or 63. Sixty are scored. The other 10 are unscored items being trialled for future forms. They are mixed in and not marked, so you can’t tell which ones they are.
- Same 120 minutes. That means less time per item. Associate drops from 2:00 to about 1:43, Developer Foundations from about 2:16 to about 1:43, and Architect Professional from about 1:54 to about 1:43. If you practised at the old pace, the pacing guide has new checkpoints.
- Scenario-based items. The new guides describe every item as scenario-based. Each item still tells you how many responses to select.
- Objectives rewritten as skills. In Associate and Architect Professional, the bullet-point objectives become named skills, each with a paragraph on what is measured. Developer Foundations already used named skills, and version 1.1 drops the weight it gave each one.
Some things stay the same: the pass mark (scaled 720), the fees, the 12-month validity and the delivery through Pearson.
Associate Foundations: seven domains become six
| Version 1.0 (until 6 Dec) | Weight | Version 1.1 (from 7 Dec) | Weight |
|---|---|---|---|
| Prompting and Task Execution | 14% | Task Definition and Prompting | 15% |
| Output Evaluation and Validation | 21% | Output Evaluation and Quality Control | 20% |
| Product and Model Selection | 12% | Capability, Model, and Context Selection | 15% |
| Workflow Integration and Solution Design | 16% | Workflow Design and Integration | 18% |
| Configuration and Knowledge Management | 12% | Configuration and Knowledge Management | 15% |
| Governance, Risk, and Responsible Use | 15% | Responsible Use, Data Protection, and Governance | 17% |
| Troubleshooting and Optimization | 10% | (no longer a separate domain) |
Troubleshooting didn’t disappear. Diagnosing a weak prompt is now a skill in Domain 1 (Prompt Diagnosis and Iteration). Working out why a configured Project has got worse is now in Domain 5 (Configuration Lifecycle and Degradation).
Three other changes are worth knowing:
- AI Fluency. The four competencies of Anthropic’s AI Fluency framework (Delegation, Description, Discernment and Diligence) are named in the guide as the basis for objectives in every domain.
- More features in scope. The guide lists Skills, connectors, memory review and incognito chats, and it adds effort level next to model tier.
- No memorized figures. The guide says items test judgment and don’t require you to recall published figures such as prices, context window sizes or rate limits.
Developer Foundations: same eight domains, new weights
| Domain | Until 6 Dec | From 7 Dec |
|---|---|---|
| Applications and Integration | 33.1% | 20% |
| Agents and Workflows | 14.7% | 17% |
| Prompt and Context Engineering | 11.0% | 15% |
| Model Selection and Optimization | 16.8% | 12% |
| Tools and MCP (Tools and MCPs in 1.0) | 10.6% | 10% |
| Eval, Testing, and Debugging | 2.6% | 10% |
| Claude Code | 3.1% | 8% |
| Security and Safety | 8.1% | 8% |
The big shift is away from API integration and towards evaluation and Claude Code, which together go from under 6% to 18%. If you were planning to skim those two domains because they barely counted, that plan no longer works. The new guide also spells out choosing between built-in tools, custom tools, Skills and MCP as a skill of its own.
Architect Professional: same seven domains, new weights
| Domain | Until 6 Dec | From 7 Dec |
|---|---|---|
| Integration | 19% | 21% |
| Solution Design and Architecture | 17% | 18% |
| Evaluation, Testing, and Optimization | 16% | 17% |
| Governance, Safety, and Risk Management | 14% | 16% |
| Claude Models, Prompting, and Context Engineering | 13% | 13% |
| Developer Productivity and Operational Enablement | 7% | 8% |
| Stakeholder Communication and Lifecycle Management | 14% | 7% |
Stakeholder communication halves. Integration and governance gain the most. The new guide also names Architect Foundations as recommended preparation, though it isn’t required.
If your date is close to 7 Dec
- Check the date and time zone in your Pearson booking, then download the guide that applies.
- For Associate, map your notes onto the six new domains before you study further. The troubleshooting skills now sit in Domains 1 and 5, so check your notes cover them there.
- Practise at the new pace if you sit on or after 7 Dec. Ten more items in the same 120 minutes is a real change of pace.
Each certdrill exam page shows both versions until 6 Dec, then only the new one. The free forms are built against the 7 Dec domains. If you choose the earlier version, the results screen tells you it is scoring against the new domains. The changes log lists every dated program change with its source.