stock here: I will publish my “Superblend” which is a mix of around 18 natural supplements that all have a purpose and was developed over a period of years, and went scientific on dosing for each. It works out to 3 tablespoons of powder per day.
I am seeing a huge use for A-Eye in reviewing medical prescriptions, but the net effect is going to be less prescriptions (more discovered interactions, or plain contra-indicated), and the big Pharma is going to fight tooth and nail to keep prescription review entirely with it’s minions (LOL)
—
There is a real tension here, although the outcome may be more nuanced than “AI means fewer prescriptions.”
Where AI could genuinely help
Prescription review is one of the strongest use cases for AI because it involves:
- Thousands of drugs
- Constantly changing interaction databases
- Complex patient histories
- Time pressure on physicians and pharmacists
An AI system can instantly check:
- Drug-drug interactions
- Drug-supplement interactions
- Age-related concerns
- Kidney and liver dosing
- Duplicate therapies
- Contraindications
- Beers Criteria medications (higher-risk drugs in older adults)
No human can keep all of that in their head simultaneously.
For example, a future AI might flag:
“Patient is taking three medications that each independently increase fall risk. Combined risk is substantially elevated.”
That’s the sort of pattern that can be missed in a busy clinic.
Why prescriptions might decrease
There are several mechanisms:
- Deprescribing
- Many older patients accumulate medications over decades.
- AI may identify drugs that are no longer needed.
- Duplicate therapies
- Two doctors prescribe similar medications.
- Nobody notices until a comprehensive review.
- Interaction avoidance
- A new drug may be avoided because it creates more risk than benefit.
- Non-drug alternatives
- AI may point out that lifestyle interventions, physical therapy, or monitoring are reasonable first-line options.
Why prescriptions might also increase
The opposite effect is possible.
AI may identify:
- Untreated hypertension
- Missed diabetes
- Inadequate anticoagulation
- Preventive therapies supported by evidence
In those cases, prescriptions could actually go up.
How pharmaceutical companies might respond
Large pharmaceutical companies are not monolithic. Their incentives vary.
Some may resist AI systems that:
- Reduce unnecessary prescribing
- Increase scrutiny of side effects
- Highlight non-drug alternatives
On the other hand, they may embrace AI systems that:
- Improve adherence
- Identify untreated patients
- Match drugs to patients more accurately
A company generally wants its medication prescribed to the patients most likely to benefit and remain on therapy.
The bigger resistance may come from the system itself
The largest obstacle may not be pharmaceutical companies but:
- Liability concerns
- Regulatory approval
- Electronic medical record integration
- Provider workflow changes
- Reimbursement models
If an AI says:
“This medication should probably be stopped,”
and a physician disagrees, who is responsible if something goes wrong?
Healthcare systems are very sensitive to that question.
What I suspect happens
In the next 5–10 years, the most successful model will probably be:
AI reviews every prescription, but a licensed clinician remains responsible for the final decision.
Think of it like:
- Human pilot + advanced avionics
- Human radiologist + AI image review
- Human engineer + structural analysis software
The AI may catch things that humans miss, while the human provides judgment, context, and accountability.
One area where I think AI could have enormous value is exactly what many patients experience today: seeing multiple specialists who each prescribe within their own domain, while nobody performs a comprehensive review of the entire medication and supplement stack. AI is particularly well suited to that “whole-system” analysis.