Rethinking Assessment in the Age of Artificial Intelligence and Multimodal Design

The future of assessment in education is being reshaped by the influence of artificial intelligence and multimodal learning. The two readings Assessment in the Age of Artificial Intelligence by Swiecki and colleagues (2022) and Multimodal Digital Classroom Assessments by Fjørtoft (2020), together explore how technology is transforming what educators value and how learning is measured. Both works call for assessment that is not only more accurate, but also more equitable and authentic.

For me as a nurse educator this topic goes beyond the classroom. In healthcare education assessment is not simply about grading or certification. It defines what it means to be competent and compassionate in a professional role. The insights from these two articles create an opportunity to rethink assessment as a human practice guided by ethics as much as by data.

Key Insights

Swiecki and colleagues (2022), explain that traditional testing is often disconnected from how people actually learn and apply knowledge. They point out that artificial intelligence can make assessment more adaptive and continuous allowing students to receive feedback in real time. Yet they also warn that automation can undermine human judgment if educators surrender too much control to algorithms.

Fjørtoft (2020) focuses on multimodal assessment where students show their learning through a mix of writing, speaking, images, and design. This approach reflects how knowledge is shared in real life. It also gives students with different communication strengths more ways to succeed. However, it introduces complexity because teachers must understand and fairly evaluate many different modes of expression.

Together these works reveal that the goal is not to choose between technology and humanity. The goal is to combine both to create assessments that are fair flexible and meaningful.

Connection to Practice

In nursing education, assessment often relies on checklists and written exams that measure memory and compliance but not critical thinking or empathy. Applying the ideas of Swiecki and Fjørtoft we can redesign assessment so that it mirrors the complexity of real care.

Artificial intelligence can support continuous learning through simulation. For instance a system can analyze how a nursing student communicates with a virtual patient and provide immediate feedback about clarity or tone. The student can then make adjustments before working with real patients. This helps bridge the gap between classroom learning and clinical practice. However the educator must interpret the results and guide the student to see the ethical and emotional dimensions behind the data.

Multimodal digital assessment can deepen this process. A student could record a short reflection about a patient scenario create a visual concept map of clinical reasoning and upload a video showing a technical skill. Each mode represents a different way of thinking and demonstrating care. This type of assessment gives a more complete view of competence while also reducing bias against students who may struggle with one single form of communication.

For me this is the future of nursing education. Assessment should not only measure what students know but how they apply knowledge with empathy, cultural awareness, and adaptability.

Critical Analysis

Both studies challenge the assumption that objectivity guarantees fairness. Swiecki and colleagues (2022), argue that when artificial intelligence is used without transparency it can reproduce bias rather than remove it. Educators may not always understand how a system is scoring performance or what data it is using. This risk is especially troubling in healthcare where empathy and communication cannot be reduced to numerical output.

Fjørtoft (2020) shows that multimodal assessment requires educators to develop new literacies. Teachers must learn to interpret meaning in sound image and gesture rather than relying only on text. This demands training time and institutional support. Without it multimodal assessment may become more performative than inclusive.

Both perspectives reveal a shared truth. Assessment reflects our values. If we value compliance the system will reward conformity. If we value reflection empathy and diversity, our assessments must be designed to nurture those qualities. The challenge is to design systems that keep the human relationship at the center of learning while still benefiting from technological support.

In nursing education this means treating every assessment as a relational act. It is not just about determining if a student passes or fails. It is about building professional identity and ethical awareness. Educators need to approach technology critically asking whether each tool strengthens or weakens the human connection that defines good care.

Advanced Critical Question

How can educators preserve human judgment and empathy when artificial intelligence and digital tools increasingly define what counts as knowledge and performance?

Next Level Solution

The answer lies in designing what I call, human guided intelligence. Educators must use technology as an assistant not a decision maker. Artificial intelligence can process data and identify learning patterns but the interpretation must remain in human hands. Swiecki and colleagues (2022), stress that educators should never delegate ethical or contextual judgment to a system. In practice this means all AI feedback must be reviewed by an instructor before being shared with students and every automated suggestion should include an explanation of how it was generated.

Educators can also build relational validity into every assessment. This means ensuring that learning evidence includes dialogue and reflection rather than just scores. A nursing instructor might use AI to identify communication gaps in a simulated patient interaction then meet with the student to discuss tone empathy and nonverbal cues. This process turns feedback into conversation and preserves professional judgment.

In multimodal environments, educators can invite students to explain the choices behind their design or presentation. By reflecting on why they used certain images or words they strengthen both self-awareness and critical thinking. Fjørtoft emphasizes that these reflections help teachers understand how meaning is constructed rather than simply displayed.

Finally educational leaders must redefine success in digital learning. Efficiency should not be the ultimate goal. Equity and growth should. Institutions can track indicators such as belonging representation and confidence rather than only speed and completion. When teachers are supported to design ethically informed assessments technology becomes a bridge to deeper understanding rather than a barrier.

The future of assessment must therefore be relational ethical and adaptive. Human judgment guided by data but rooted in empathy is what will keep learning authentic in an age of automation.

Conclusion

Assessment is not only a measurement of learning but a reflection of how we see human potential. The studies by Swiecki and Fjørtoft remind us that true innovation balances intelligence with compassion. Artificial intelligence and multimodal design can expand access and creativity only if educators remain the moral compass of the process.

In nursing education, this vision matters deeply. Technology can support reflection feedback and fairness but only educators can teach the meaning behind care. When assessment is designed with equity curiosity and empathy at its heart it prepares both students and teachers to lead with humanity in a digital world.

Personal Reflection

Therefore, how can educators ensure that artificial intelligence and multimodal assessment tools promote equity rather than deepen existing inequalities in education and professional training?

Equity is often the first promise and the first casualty of innovation. While artificial intelligence and multimodal assessments claim to expand access and opportunity, they can easily reproduce privilege if educators do not critically examine who designs the systems, who benefits from them, and who is left out. The study by Swiecki et al. (2022) makes this tension clear. They explain that many AI tools are trained on datasets that reflect historical patterns of success. If those data privilege certain linguistic, cultural, or socioeconomic groups, then the system will continue to reward those same traits. What looks like efficiency, is often bias at scale.

Fjørtoft (2020) highlights a related concern. Multimodal assessment offers diverse ways to demonstrate knowledge, but only if students have access to the necessary tools and literacies. For example, a student who has access to design software, a quiet recording space, and reliable internet will naturally produce higher-quality multimodal artifacts than one who lacks those resources. Without equitable access and support, multimodal design risks reinforcing the digital divide it intends to bridge.

To address this, educators must view technology through the lens of design equity. Design equity means intentionally creating conditions that allow all learners to participate meaningfully, regardless of their prior experience or resources. In practical terms, this involves offering low-bandwidth alternatives, accessible templates, and flexible timelines. It also requires explicit teaching of multimodal composition skills so that assessment measures thinking and creativity rather than technical mastery.

Institutions also have a moral obligation to ensure transparency in their technology adoption processes. Before approving any AI-driven platform, decision-makers should ask whether the system provides evidence of fairness, explainability, and accessibility. If an algorithm cannot justify its decisions in plain language, it has no place in education. Transparency not only builds trust but also supports educators in aligning AI practices with professional ethics.

As a nurse educator, I believe equity is not just an educational issue, but a matter of care ethics. Nursing itself is grounded in fairness, advocacy, and relational accountability. Therefore, the same values must guide how we assess learning. A student’s ability to demonstrate compassion, reflection, or cultural awareness should not depend on their familiarity with software or digital media. Instead, technology must bend toward the learner.

One effective approach is to integrate peer and community review into digital assessment. When students review each other’s multimodal projects or AI feedback summaries, they begin to see how culture and context shape interpretation. These collaborative reviews turn assessment into a shared learning space and help educators identify inequities early. This approach mirrors collaborative care in nursing, where diverse perspectives improve both diagnosis and treatment.

Finally, the measure of equity in assessment should not be limited to outcomes but extended to experience. Educators can collect qualitative feedback from students about how supported and represented they feel during the learning process. Equity, after all, is not only about fairness in grading but also about dignity in learning.

Artificial intelligence and multimodal tools hold transformative potential, but their power must be guided by human ethics. According to Swiecki et al. (2022), technology must enhance, not replace, the wisdom of educators. When we combine that wisdom with the inclusivity envisioned by Fjørtoft (2020), assessment becomes more than evaluation. It becomes a mirror of the values we want to live by as professionals and as people.

The real test of innovation is not what it measures but what it makes possible. If we use these tools to strengthen equity, empathy, and understanding, we can turn assessment into a model of the future we want to create. One that is intelligent, inclusive, and profoundly human.

References

Fjørtoft H 2020 Multimodal digital classroom assessments Computers and Education 152 103892 https://doi.org/10.1016/j.compedu.2020.103892

Swiecki Z Khosravi H Chen G Martinez Maldonado R Lodge J M Milligan S Selwyn N and Gašević D 2022 Assessment in the age of artificial intelligence Computers and Education Artificial Intelligence 3 100075 https://doi.org/10.1016/j.caeai.2022.100075