The growing reliance on online reviews for decision-making has significantly influenced the healthcare industry.
Patients are increasingly turning to platforms like Google Reviews, Yelp, and other review sites to guide their choice of hospitals, dental clinics, and other healthcare providers.
These reviews shape perceptions of care quality, professionalism, and trustworthiness.
This study aims to assess the prevalence of AI-generated reviews in hospitals and dental clinics, identifying trends and patterns that may reveal vulnerabilities in the current system.
Using AI detection tools and datasets of reviews scraped from major platforms, the research will analyze the proportion of reviews likely generated by AI in these settings.
By understanding how and where AI-generated reviews proliferate, this study seeks to offer insights into mitigating their impact, developing better detection systems, and fostering a more transparent online ecosystem for healthcare consumers.
Ultimately, the findings of this research have broader implications for trust in digital content and the ethical responsibilities of review platforms.
The emergence of AI-generated reviews poses a new and complex challenge to this landscape.
By leveraging advancements in natural language processing (NLP) and generative AI, businesses or malicious actors can easily create convincing reviews that are indistinguishable from those written by real patients.
The implications of AI-generated reviews in healthcare settings are particularly profound. Trust is paramount in healthcare, considering that patients often make critical decisions based on online information such as reviews.
Fake reviews — whether overly positive or negative — can distort the reality of care quality.
Some potential implications of fake AI healthcare reviews include:
As AI tools become more accessible, policymakers, healthcare providers, and platform administrators must work together to ensure that the reviews guiding patients remain authentic and reliable.
Our study analyzed the prevalence of AI-generated reviews in key healthcare sectors in the US and Canada including:
Let’s take a closer look at which healthcare sectors exhibit AI-generated reviews:
In our analysis, we found that American hospitals exhibited a 7% AI-generated review rate.
Further, the regions with the highest presence of AI hospital reviews were:
The presence of reviews that were Likely AI in Canadian hospitals was notably higher at 12.1%.
The Canadian regions with the highest percentage of AI reviews were:
Our study found that US dental clinics had a higher percentage of reviews that were Likely AI than in a hospital setting.
13.1% of reviews in dental clinics were Likely AI in contrast to 7.1% of hospital reviews.
The US regions with the highest percentage of AI reviews in dental clinics are:
Similar to the comparative findings of US hospitals and dental clinics — Canadian dental clinics also had a significantly higher percentage of AI reviews than Canadian hospitals.
20.7% of reviews for Canadian dental clinics were Likely AI vs. 12.1% of AI reviews for Canadian hospitals.
Dental clinics had the highest rate of AI reviews in a Canadian healthcare setting.
In Canada, the top regions with the highest percentage of AI dental clinic reviews were:
The highest overall percentage of AI reviews in the US healthcare sector was present in plastic surgery clinics with a total of 28.9% of the reviews detected as Likely AI.
The US regions with the highest percentage of AI reviews in plastic surgery clinics were:
In contrast to the US, which saw the highest proportion of AI reviews in plastic surgery clinics, in Canada the highest percentage was in dental clinics (20.7%), notably higher than in Canadian plastic surgery clinics at 17%.
The region in Canada with the highest percentage of AI reviews in plastic surgery clinics was Toronto, located in the province of Ontario.
The analysis revealed significant differences in the prevalence of AI-generated reviews across healthcare settings in the United States and Canada.
These findings suggest varying vulnerabilities to AI-generated reviews within the healthcare sector, influenced by both the type of institution and regional factors.
The disparity between the U.S. and Canadian rates also raises interesting questions about regional differences in review ecosystems.
The higher AI-generated review rates in Canada, particularly in dental clinics, could indicate differences in regulatory oversight, platform algorithms, or cultural attitudes toward online reviews.
For example, less stringent oversight on review authenticity or a greater reliance on review platforms for healthcare decision-making might contribute to these elevated rates.
The relatively higher prevalence of AI-generated reviews in dental clinics compared to hospitals in both countries could reflect differences in market dynamics and patient interaction.
Dental clinics often operate in highly competitive local markets, where reviews play a critical role in attracting patients. The increased use of AI to generate reviews may stem from efforts to enhance visibility or manage reputations in this competitive landscape.
Conversely, hospitals, which tend to have a more stable patient base and rely on broader institutional trust, may face less incentive or pressure to manipulate reviews.
These findings underscore the risks posed by AI-generated reviews in healthcare, a sector where trust and reliability are paramount.
Fake reviews, whether positive or negative, can distort perceptions of care quality, potentially leading to suboptimal patient decisions.
For example, a dental clinic with predominantly AI-generated positive reviews might attract patients under false pretenses, risking disappointment or even harm if the clinic fails to meet expectations.
Fake reviews also pose serious implications from the perspective of healthcare providers.
Legitimate healthcare providers could have their clinic and medical reputation unjustly damaged if fake reviews contain incorrect negative information.
The higher rates observed in Canada, particularly for dental clinics, and in the US, particularly for plastic surgery clinics, suggest an urgent need for platforms to enhance their AI detection mechanisms. Additionally, it highlights a need for policymakers to consider guidelines to mitigate the spread of misleading content.
Further, the disparity between hospitals and dental clinics also highlights the importance of tailoring interventions to specific healthcare contexts, as the drivers and impacts of AI-generated reviews may differ between these settings.
Incorporating a reliable AI detector, like Originality.ai is key in identifying the presence of AI-generated healthcare reviews. Originality.ai has exceptional AI detection accuracy as established through our AI Detection Accuracy Study and through third-party AI-detection studies.
Addressing the prevalence of AI-generated reviews requires a multi-faceted approach. Review platforms must invest in robust AI detection to flag and mitigate fake content.
Healthcare providers should be proactive in encouraging genuine reviews from patients to dilute the influence of AI-generated content.
For policymakers, these findings highlight the need for regulations that prioritize transparency and accountability in online review systems.
Ultimately, the integrity of online reviews in healthcare is critical not only for individual patient decisions but also for the broader trust in digital content.
By understanding the patterns and implications of AI-generated reviews, stakeholders can work together to create a more trustworthy and reliable online ecosystem for healthcare consumers.
This study analyzed hospital, dental clinic, and plastic surgery clinic reviews for AI-generated content using a two-step process:
Data Collection: Reviews were gathered via the Google Places API, focusing on text, timestamps, and metadata (hospital/clinic name, city, state). Up to 50 reviews per location were stored in CSV files, with delays added to respect API limits.
AI Detection: Reviews were analyzed with the Originality.ai API, excluding those under 50 words. Results included AI likelihood scores and binary classifications (AI-generated or not).
Challenges included API rate limits and language limitations. The study highlights trends in AI-generated reviews and offers insights for improving trust in online content, particularly in reputation-sensitive healthcare sectors like plastic surgery.
We tested Originality.ai using the same dataset and methods of the “ESPERANTO: Evaluating Synthesized Phrases to Enhance Robustness in AI Detection for Text Origination" study accessible via Cornell University. Here's how it stands out as a leader in the domain.
