AI in the Consulting Room: Hype vs Reality for Independents

Patients are arriving at your consulting room having already consulted ChatGPT. Your OCT machine is flagging findings before you've sat down. And the pre-test room auto-refractor now comes with an onboard AI confidence score. This is not the future of independent optometry. It is the present. The question is not whether AI belongs in your practice — it is which parts are genuinely useful and which are expensive noise.

The word AI gets attached to everything right now. Recall software. Diagnostic imaging. Appointment booking. Frame recommendation engines. Most of it is automation with a marketing budget. Some of it is genuinely transformative. Independent practice owners do not have time to audit every product claim — so here is an honest look at what AI is actually doing in clinical optometry in 2026, where the evidence is solid, and where to be cautious.

The patient who already Googled their symptoms — and then asked ChatGPT

This is the most immediate shift. Patients are arriving in consulting rooms after using large language model tools such as ChatGPT or Google Gemini to research symptoms before their appointment. They have a list. They have a theory. Sometimes they have printed something out.

For a corporate multiple running 20-minute appointment slots, this is a problem. There is no time to untangle a patient's self-diagnosis within a standardised workflow. For an independent practice with longer appointment structures and a genuine dispensing conversation, it is an opportunity.

An informed patient is an engaged patient. The independent who treats that pre-consultation research as a starting point — rather than an interruption — is in a stronger position than one who dismisses it. Know what these tools are telling patients about common conditions. Glaucoma risk. Dry eye. Macular degeneration. The information is broadly accurate at the population level and lacks the clinical nuance your examination provides. That gap is where you add value.

OCT with AI analysis: where the evidence actually sits

This is the area where AI in optometry has the most robust clinical foundation. OCT manufacturers, including Topcon, Heidelberg and Zeiss, have been building AI-assisted layer analysis and pathology flagging into their platforms for several years. The technology is real. The clinical utility is real. The peer-reviewed evidence is solid.

AI-assisted OCT has demonstrated the ability to detect and classify multiple retinal conditions — including age-related macular degeneration, diabetic retinopathy and glaucoma — at accuracy levels comparable to specialist review. A 2022 community screening study published in Frontiers in Cell and Developmental Biology evaluated AI-assisted OCT detection across 15 distinct retinal disorders, finding the approach both accurate and feasible in non-hospital settings (Bai et al., Frontiers in Cell and Developmental Biology, 2022).

Platforms like Altris AI offer cloud-based OCT analysis to help practitioners review macular imaging and flag findings that warrant further investigation or referral. Practitioners using this kind of tool cite two specific benefits: a reduction in the risk of missed findings, and a stronger clinical conversation with patients who can see their prognosis visualised.

What AI-assisted OCT does not do is replace the clinician. It flags. It scores confidence. It surfaces findings faster. The decision — the clinical judgement, the referral, the management plan — remains yours. That is not a limitation. That is the correct division of labour.

For independent practices already running OCT as a clinical service, checking whether your current platform includes AI-assisted analysis — and whether it is switched on — is worth doing this week.

Auto-refraction with AI confidence scoring: useful or distracting?

Pre-test room auto-refractors with onboard AI confidence scoring are now appearing in equipment catalogues. The pitch is that the system flags when its own result is less reliable — directing the clinician to pay closer attention before finalising the subjective refraction.

In principle, this is sensible. Auto-refraction has always been a starting point, not a conclusion. A system that quantifies its own uncertainty is more honest than one that outputs a result without qualification.

In practice, the value depends on how well the confidence scoring is calibrated and whether it genuinely changes clinical behaviour, or simply adds a layer of information that experienced clinicians already account for implicitly. If you are considering new pre-test equipment that includes this feature, ask the manufacturer for clinical validation data specific to the scoring algorithm. Not marketing materials. Validation data.

Where the hype outruns the evidence

AI-powered frame recommendation engines. AI-generated recall messaging that adapts to patient behaviour patterns. AI appointment scheduling that predicts no-shows. All real products. All are currently being marketed to independent practices.

Some of this is useful automation dressed up in the language of AI. Recall systems that send different messages based on patient response history are not artificial intelligence in any meaningful clinical sense — they are conditional logic. That is fine. Conditional logic can be valuable. Just do not pay AI prices for it.

The benchmark question for any AI-marketed product aimed at an independent practice is straightforward: what does this system do that your current process cannot do, and is there evidence that the outcome is better for patients or the practice? If the answer requires a lot of words to arrive at something vague, the product is probably not ready yet.

The independent advantage here is real — if you use it correctly

Corporate multiples are investing in AI at scale. Automated triage. Centralised image analysis. Standardised AI-assisted workflows pushed down to every branch. That investment will make the corporate model more efficient. It will not make it more personal.

The independent practice that uses AI-assisted OCT analysis to support better clinical conversations — showing patients their retinal health, explaining what the system flagged and why — is using the same technology to deliver something multiple cannot replicate at scale: a genuine one-to-one clinical relationship, with a practitioner who knows the patient and can contextualise the data.

AI does not commoditise independent optometry. It raises the floor on what good looks like. Independents who are already above that floor — clinically, relationally, in terms of appointment time and personal continuity — benefit from the tools without being threatened by them.

The independents who should be paying attention are those whose competitive advantage has been proximity and convenience, not clinical depth. If the only reason patients choose you over the multiple down the road is that you are closer, AI-enhanced corporate efficiency will close that gap. Clinical specialism and genuine patient relationships will not be replaced by it.

Three things to do this month

First, check your OCT platform. Does it include AI-assisted analysis? If so, are you using it, and are you using the output in your patient consultation? If not, speak to your supplier about what is available on your current system before investing in anything new.

Second, read what ChatGPT and similar tools currently say about the three most common conditions your patients ask you about. Dry eye. Glaucoma risk factors—macular health. Know the information patients are arriving with. Build it into your pre-consultation process rather than discovering it mid-appointment.

Third, when evaluating any new AI-marketed product, ask for validation data. Not a case study. Not a testimonial from another practice. Published evidence or independent validation of the specific claim being made. If the supplier cannot provide it, the product is not ready for clinical deployment.

AI is in your consulting room already, whether you have made a deliberate decision about it or not. The question is whether you are using it intentionally — or letting it use you.

If you are building or growing an independent practice and want to talk through how technology investment decisions fit into your wider business strategy, this is exactly what our Grow Independent service is designed for. Book a Free 20-Minute Practice Growth Call