Cognitive Partnership
The phrase “AI will assist, not replace” is now commonly used across health care, including in optometry, to reassure clinicians about the role of artificial intelligence in patient care. Though well-intentioned, it has become a kind of rhetorical comfort blanket, repeated so often that it obscures the deeper transformation underway. One of the key limitations of this phrase is that it frames AI as a passive tool, like a phoropter or autorefractor. These devices, while valuable, have no agency and rely entirely on human direction to function. AI systems, by contrast, are increasingly ambient, working in the background to analyze data, detect patterns and generate insights without prompting. This quiet but continuous presence is not merely changing what eye care professionals (ECPS) do. It is beginning to reshape how they think, reason and make clinical decisions.
Traditional clinical reasoning in optometry has long relied on intuitive pattern recognition, tacit knowledge and experiential judgment. These are all modes of thought shaped by what the clinician sees, hears and recalls in the moment. AI introduces a parallel stream of cognition, grounded in statistical inference, predictive modeling and vast cross-patient comparisons. Clinicians are learning to interpret machine-generated insights. As such, clinical reasoning is becoming a dynamic dialogue between human and machine. This evolution demands not only a redefinition of clinical expertise, but a reimagining of how ECPs are trained for the future of care.
A New Set of Skills
In this new health care milieu, practitioners are no longer the sole interpreters of diagnostic cues,. They are participants in an ongoing feedback loop with intelligent systems. ECPs will need new technical skills to understand, assess and effectively engage with AI in clinical practice. While they will not need to be data scientists or computer engineers, ECPs will need to expand their data proficiency beyond traditional research literacy. They will need skills in interpreting probabilistic outputs, understanding how AI systems express uncertainty, recognizing bias and questioning how the system makes decisions. They will also need to be able to evaluate the reliability of an AI system’s answers, and whether the system’s level of confidence reflects real-world accuracy. This is especially important since these systems often provide results without explaining how they arrived at them.
A New Mindset
Working effectively with an AI partner will also require a significant shift in the clinician’s mindset. Many clinicians are already beginning to express discomfort as AI systems outperform human judgment in specific domains, particularly in areas like retinal image analysis, early disease detection or triage-level pattern recognition.
When an algorithm consistently identifies pathology more quickly or more accurately, it challenges long-held notions of clinical expertise. For some, this can provoke feelings of self-doubt, defensiveness or a sense that their professional identity is being undermined. The discomfort is rooted in what it means to be a skilled clinician when diagnostic mastery is no longer a uniquely human trait. To work effectively in this new paradigm, clinicians will need to shift their perception of expertise, from sole authority to collaborative decision-makers who integrate both human judgment and machine insight. This evolution doesn’t diminish their role. It elevates it, enabling them to offer more accurate, informed and patient-centered care than ever before.
Metacognition
A crucial component of this evolving mindset is metacognition: the ability to reflect on one’s own thinking. Practitioners must be trained not only to interpret AI outputs, but to examine how their clinical judgment is influenced by those outputs. This includes knowing when to trust, when to question and when to reconsider their conclusions without ego or defensiveness.
Consider an AI suggesting a treatment plan that is very different from the ECP’s preferred choice. Should the clinician defer to the algorithm, rely on their instincts or seek a middle ground? In such moments, the challenge is not only to evaluate the AI’s recommendation, but to critically assess one’s own reasoning. This kind of metacognitive awareness will be essential for clinicians to navigate AI-augmented care with confidence, clarity and ethical integrity. Educating future ECPs must therefore include critical thinking and data literacy, as well as intentional support for the emotional and cognitive shifts required to thrive in a collaborative, AI-driven paradigm.
Human Skills
As clinical reasoning becomes shared with machines, our understanding of the clinician’s role in patient care must also adapt. There are distinctly human aspects of care—empathy, compassion, presence, storytelling and trust-building. These are essential to the evolving identity and purpose of the modern practitioner in an increasingly automated clinical landscape. A skilled clinician must bring more than knowledge to the room. They must bring emotional instinct, detecting unspoken concerns and anticipating distress. They must craft a plan of care that aligns not just with the diagnosis, but with the person. While AI can deliver probabilities, only a human can read the subtle tension in a patient’s posture, recognize when fear clouds comprehension and respond with empathy that turns confusion into trust.
How Are We Training Students?
To prepare students for this future, optometric education must prioritize the development of relational and communication skills with the same rigor historically reserved for anatomy, pathology and pharmacology. Role-playing exercises, case studies and patient interaction simulations have long been part of clinical training. However, they have traditionally played a supporting role, supplementing, rather than equaling, the weight of technical knowledge. In an AI-augmented environment, these practices must take center stage.
This means training students not just to speak clearly, but to listen deeply; not just to inform, but to interpret. For example, when an AI flags an early-stage disease risk, it falls to the practitioner to explain the nuance, to reassure without downplaying its gravity, to prepare without alarming. Similarly, when multiple treatments are possible, clinicians guide patients in shared decisions, balancing predictive data with individual preferences and lifestyle.
These conversations require advanced skills in emotional intelligence, cross-cultural communication and narrative framing. A patient from a medically underserved background may bring a different level of trust, health literacy or skepticism to the conversation. A patient with chronic conditions may be overwhelmed by information. In both cases, the clinician must act as a translator. They must take data-rich insights and turning them into personalized, understandable and compassionate care plans.
In a hybrid clinical world, where part of the work is done by algorithms and part by human beings, what sets the ECP apart will not be raw analytical power, but their ability to connect, contextualize and care. These skills are critical and they must be treated as such in the future of optometric education.
Call to Action
The phrase “AI will assist, not replace” no longer captures the complexity of what’s unfolding. AI is no longer just a tool—it is becoming a constant presence in the exam room, transforming how care is delivered and how decisions are made. It changes not only what ECPs do, but who they must become. This moment demands a bold, strategic response from institutions of optometric education.
Curricula
First, curricula should be redesigned to meet emerging realities. AI should not be siloed as a niche topic. It should be integrated as a foundational thread across clinical education, from diagnosis and decision-making to ethics and communication. This includes training in AI literacy, probabilistic reasoning and collaborative decision-making, alongside an elevated emphasis on human-centered skills like empathy, interpretation and trust-building.
Learning Environments
Second, institutions should invest in immersive, future-ready learning environments, creating labs for real-time AI interaction. Simulations should be expanded to help students translate algorithmic insights into patient-centered dialogue. This will redefine the clinical encounter as a three-way partnership between patient, provider and machine. Students need space to practice that complexity.
Prepare Faculty
Third, equip faculty to lead the transformation. Educators cannot prepare students for a future they are not prepared to navigate themselves. Administrators should prioritize giving faculty time, training and support to explore new technologies. This will allow them to evolve their teaching methods and help shape the next generation of clinical reasoning and care.
The future of clinical care is already taking shape around us. Only by moving deliberately and decisively can institutions position themselves to thrive in this new ecosystem. However, that requires an honest assessment of current practice and a willingness to invest in innovation. Institutions that act now, redefining curricula, empowering faculty and embracing innovation, will be poised to shape the future of optometric care.
Authors
