AI Build and AI Interpret are now available
We’re thrilled to announce the release of two exciting new AI-powered features in Brilliant Assessments: AI Build and AI Interpret. These tools are designed to dramatically accelerate the creation of assessments and add significant value to the delivery of feedback.
Why AI, and why now?
AI is everywhere, but that doesn’t mean it should be everywhere. We didn’t rush to release AI just to check a box. Instead, we spent over a year designing features that genuinely support how professional assessments are built, scored, and delivered.
Brilliant Assessments is used to create complex, flexible, and often very detailed, expertise-driven assessments. So any AI we introduced had to meet that bar. It had to be really, really useful, not hype-based and ‘trendy’.
A brilliant assessment, pun intended, still requires thought, experience, and context. Our AI features don’t replace your expertise; they complement it. You bring the insight and direction, as someone who likely knows more about your subject matter than AI does. What AI provides are shortcuts, such as writing questions, drafting feedback, or analyzing responses in ways that previously only a human assessor could manage.
AI in Brilliant Assessments helps you complete high-quality work more efficiently, uncover new perspectives, and create opportunities for growth and scalability.
Also, as we mentioned, there are two new AI features. And while they might seem similar at first, because of the technology they utilize, they are actually on completely different levels. So let’s get into it…
AI Build: A quick-start tool for assessment building
AI Build is the tool many expected when we announced AI was coming. It helps you build your assessment by generating sections, subsections, questions, and feedback text based on your topic and prompts. AI Build is included in all plans with no limits and no extra charges.
We expect this feature to be most valuable for those building their first assessment or creating something quickly on a trial account to see if Brilliant Assessments is a good fit. Creating a quality assessment from scratch takes time. You need to plan your sections, write relevant questions, and prepare feedback for different score bands. AI Build speeds this up by generating structured, ready-to-use content based on your prompts, settings, and description of the assessment.
AI Build starts with your assessment title and description. These two fields do most of the heavy lifting when it comes to context. A clear, specific title gives the AI a sense of what kind of content to generate. A strong description adds more detail about who the assessment is for, what it’s measuring, and what kind of tone or outcome you are aiming for.
Once that basic context is in place, AI Build can successfully provide:
Sections that reflect the key themes of your assessment
Subsections that go into specific areas or skill sets
Questions that match your selected answer types
Draft report text for each answer option or score range/rating
You can also adjust how the content is written. Want something formal or professional? Prefer a conversational or coaching tone? Need a short summary or a full paragraph? You can include all of that in your prompt, and AI Build will adjust the output accordingly.
But if you want to learn how to prompt like a pro, we’ve pulled together a pretty detailed how-to guide here.
AI Interpr: Enterprise-grade feedback, designed by you, powered by AI
AI Interpret is the tool that takes things to the next level, stepping in once responses start coming in. It is designed for organizations delivering high-touch, high-value assessments, where manually interpreting written responses, documents, and qualitative input would normally take hours. This is where scale and personalization come together.
AI Interpret brings enterprise-grade intelligence to your feedback and scoring process. It transforms open-text answers, uploaded files, and combined inputs into structured, high-quality content you can immediately use in your reports.
Here’s what AI Interpret can do:
Write detailed, personalised feedback for each respondent
You can prompt AI to read open-ended answers and produce feedback that reflects your preferred tone, format, and level of depth (eg, short summaries, bullet points, or full narrative paragraphs)
Interpret uploaded files and written answers together
AI can read and respond to attached PDFs, strategy documents, audit reports, or spreadsheets, combining them with written answers to produce more complete feedback.
Score qualitative input based on your own criteria
You define the scale and scoring rules. For example, you can instruct the AI to assign scores from 0 to 5 based on relevance, tone, or alignment with a specific model.
Summarise themes across teams or cohorts
AI Interpret can be used to identify shared strengths, gaps, or trends across a group of responses, enabling you to deliver group-level feedback or insights without assessor analysis.
Match your reporting style and structure
With your prompt, you can maintain control over tone, format, and layout. Ask for one sentence or a full paragraph. Keep it formal or use a coaching tone.
Insert directly into your existing reports
AI Interpret can be added to reports using a single answer text merge string or pulled into your charts using segmentations for each score.
To help you get even more out of AI Interpret, check out our detailed AI Interpret prompt guide. It includes real examples, formatting tips, and suggestions for different use cases — from individual feedback to cohort-level summaries and scoring logic.
Built with Privacy and Security in Mind
We understand that when you're using AI for assessments (especially when dealing with sensitive client, business, or strategic data), security is non-negotiable. We’ve built AI into Brilliant Assessments with the same rigorous standards you expect from a professional-grade assessment platform, so you can confidently innovate without compromising security or privacy.
Here’s how we keep your data safe:
We use a secure model
We utilize Microsoft Azure’s Semantic Kernel to connect to OpenAI’s GPT o4 model. This is an industry-leading AI solution that strikes a balance between power, reliability, and compliance.
Your data is never stored or shared
When you use AI features in Brilliant Assessments:
No data is stored on Microsoft Azure or OpenAI servers.
No prompts, responses, or uploaded documents are ever used to train OpenAI models.
Your proprietary content stays your own.
This ensures your intellectual property, customer information, and assessment logic remain entirely confidential.
Secure Storage and Context Management
We store prompt history securely within your Brilliant Assessments account to maintain context across interactions. This makes AI responses more coherent, especially for follow-up prompts or iterative suggestions.
All data is stored in our own secure AWS infrastructure and is never exposed to external platforms beyond what’s needed for real-time inference.
End-to-End Encryption
All data transmitted between our AWS environment and Microsoft Azure is fully encrypted in transit. This ensures that no information can be intercepted or read during processing, even during cross-cloud communication.
AI and the Brilliant Assessments Subscription Plans
AI Build and AI Interpret are available for free to all users on all plans.
See how it works or try it out
AI Build and AI Interpret are now available in Brilliant Assessments to help you build faster, review responses more efficiently, and deliver high-quality feedback at scale, all while keeping full control over the output.
Next steps to getting started
Book a demo with our team
Contact support to start your trial.
Read our guides for AI Build and AI Interpret.
Send us your feedback—we’d love to hear how these tools help you save time and improve quality, or if you have any ideas on how we can make the tools even better.