
assessmentQ answers: possibilities and limitations of AI in e-assessment
AI is rapidly reshaping entire industries by streamlining processes and improving efficiency. And e-assessment is no exception. With new AI models and tools popping up all the time, it can feel overwhelming to know what is actually useful. That is why we have put together this clear overview of how AI fits into e-assessment. You will get a better sense of what it can do, where it helps, where it doesn’t, and why finding the right balance matters.
AI in e-assessment
AI is making big moves in EdTech, particularly in the realm of e-assessment. From creating questions to enhancing feedback and proctoring, its possibilities are impressive. But they come with challenges too. The key? Finding the right balance between where AI can help, but still keeping people in control.
AI supports people, it does not replace them
Let’s start with question creation. AI can generate questions based on things like syllabi, learning goals, or existing item banks. That is helpful. Don’t rely on it blindly, though. Not every AI-generated question is exam-ready. You still need people to check for clarity, relevance, and fairness.
After the exam, AI can help review the quality of your questions. It can spot patterns, flag missing topics, and catch readability or bias issues. But again, this works best when paired with human judgment.
For this, transparency is essential. According to the EU AI Act, all AI-generated content must be clearly labelled. People have a right to know when they are looking at AI-generated material.
Making assessments fairer and more accessible
AI can make assessments more inclusive too. Features like adaptive testing or text-to-speech functionality can tailor the exam experience to individual candidates’ needs. But be careful: if not used thoughtfully, these tools can introduce new bias or unfair advantages.
Intelligent proctoring, with limitations
In high-stakes environments, AI-powered proctoring tools are increasingly common. It can monitor behavior, flag suspicious activity, and save reviewer considerable time and resources. At the same time, they raise concerns around transparency, data privacy, and bias.
The EU AI Act treats AI-based proctoring as a high-risk use. That means that it must be explainable, transparent, and overseen by people. Choosing a platform that adheres to these regulations is therefore a necessity.
Helping with grading and feedback
AI can assist with grading both closed and open-ended questions. It has been automatically scoring multiple-choice questions for years, but newer tools can now evaluate open questions as well. It checks for structure, grammar, and how well a student answered the actual question. This can dramatically reduce the grading time. It is even possible for AI to automatically write feedback.
But there is a catch: the EU AI Act flags the use of AI for evaluating purposes as high risk, meaning that it is forbidden to let AI give out scores. In fact, all educational applications of AI are deemed high-risk systems by the EU AI Act. So use AI as a smart assistant rather than as a replacement for your colleagues or employees.
Possibilities
- Creating and refining questions
- Reviewing proctoring footage more efficiently
- Spotting trends across student performance
- Providing scalable, personalised feedback
Limitations
- Data privacy and compliance: platforms must meet GDPR and EU AI Act standards.
- Overreliance: AI should support decisions, not make them.
- Bias: Poor-quality data leads to biased. Diverse, reliable data is a must.
How we use AI in assessmentQ
At assessmentQ, we are excited about what AI can offer, but at the same time we are careful about how we use it. We do acknowledge the benefits and challenges and that is why we only introduce new features that really add value to our customers. We want to make sure that everything we develop will help the user and not pose any challenges.
Every AI-based feature is clearly marked with an icon, showing transparency when using AI.
Here are a few of the smart AI-based features that can assist you in creating better tests and exams.
Automatic distractor generation
Struggling to come up with good wrong answers? That is where our ‘distractor generation’ feature can come in handy. AI generates five plausible distractors for your multiple-choice questions, saving you time and effort. You can even refine the generated distractors according to your needs.
Conjugate verbs
An item type that is fully AI-powered. AI scans your text, detects all verbs, and turns them into fill-in-the-blank questions. Candidates complete these blanks with the correct verb form. Great for testing grammar skills in English and French.
AI-powered PDF import
Need to move questions from a PDF into assessmentQ? This AI feature can help you with that. It lets you upload your PDF file and select the text per question. In just a few clicks, your multiple-choice questions, open questions, and ‘fill gaps in text’ questions are imported.
AI-based flags in the proctoring integration with assessmentQ
When using proctoring, the seamlessly integrated proctoring solutions in assessmentQ also provide AI-based assistance. Proctors can review the recorded exam sessions directly in assessmentQ. The AI automatically flags suspicious incidents or events. This way the proctors do not have to review everything from start to finish. It saves them time and helps you focus on what really matters.
Conclusion
AI is transforming the landscape of e-assessment. It can make exams more efficient, fairer, and easier to manage. From creating questions to ensuring validity and saving time for evaluation, its potential is immense. But it is important to be mindful of the challenges. The EU AI Act reminds us that educational applications of AI are high-risk and must be approached with transparency, responsibility, and human oversight.
At Televic Education, we embrace AI’s possibilities without losing sight of its limitations. By only introducing features that have a genuine benefit, and clearly labelling AI-generated content, we ensure that innovation serves our clients in a meaningful and compliant way. The future of e-assessment is not about replacing people with AI but empowering them through it.
