Not long ago, if dentists wanted answers to clinical questions, they generally knew where to look. Research findings moved through a relatively predictable pathway, through sources such as PubMed or Google Scholar.

Studies were conducted, journals published the results, professional organizations evaluated emerging evidence, and thought leaders translated the science into practical relevance. The diffusion of innovation by communication and networks was predictably slow, with a chasm between the innovators and early adopters. Over years, even decades, clinicians decided what deserved a place in everyday practice.

The process was never perfect. It could be slow, occasionally political, and sometimes frustratingly resistant to change. The rules for deciding which types of evidence met the criteria for inclusion varied. Still, it worked reasonably well because scientific authority and information authority tended to move together.

Today, those pathways are beginning to shift. Increasingly, patients are arriving in practices after spending time with ChatGPT, AI-enhanced search platforms, YouTube videos, social communities, and online forums.

Clinicians themselves are experimenting with AI for diagnostic support, digital workflow, continuing education, and content development.¹⁻⁴ Information no longer simply travels from journal publication to conference lecture to slow clinical adoption.

It now moves at speed and scale, through an increasingly complex network of algorithms, summaries, citations, semantic relationships, and recommendation systems that help determine which ideas become visible and which quietly fade into the background.

That shift raises a question dentistry has barely begun to discuss: Can evidence-based dentistry survive the AI search era?

To be clear, this is not an argument that science suddenly matters less. Quite the opposite. The issue is whether evidence alone still guarantees influence in a world increasingly shaped by discoverability – which like traditional SEO, can be manipulated. The criteria for which type evidence to use and the methods must be evaluated as well. Gold standard evidence is clinical trials, yet dentistry has failed to include oral health, even one question, in these well-funded studies.

For decades, academia operated under a familiar pressure: publish or perish. Researchers who contributed to the literature advanced careers, built reputations, and influenced the direction of the profession. Since the government’s shift away from university support as centers for innovation, evaluation of relevant evidence is also shifting.

Today’s AI-powered search may be introducing a different reality, one where information that is digitally visible, repeatedly cited, linked, discussed, translated, and surfaced across multiple channels gains influence beyond traditional publication models. Instead of taking decades for adoption, the chasm is closing between innovators and majority adopters.

That raises an uncomfortable possibility. Are we moving into an era of “publish or disappear”?

Organizations that fail to actively contribute evidence-based information into the digital ecosystem may gradually lose influence over what clinicians and patients ultimately encounter. Accessibility will be key, as traditional paywalls of scientific publications, profitable corporate entities, themselves barriers to diffusion of innovation. 

That observation is not intended as criticism of scientific rigor. It may simply reflect the reality of how information ecosystems increasingly operate.

Artificial intelligence systems do not evaluate evidence in the same way clinicians do. They do not independently determine scientific truth, or the best evidence to use. Instead, they are algorithm derives, and they recognize patterns and relationships across enormous collections of all types of information.

Researchers have increasingly pointed out that AI systems rely on signals such as citations, semantic associations, repetition, authority indicators, and contextual relationships while requiring safeguards around transparency, oversight, and evidence quality.⁵⁻⁸

That distinction matters because AI may not naturally reward the strongest science. Under certain conditions, it may reward the most visible science.

The concern here is not necessarily misinformation, although that remains important. The larger issue may be information imbalance. Evidence-based organizations often move carefully for good reasons. Peer review takes time. Consensus development takes time. Responsible science takes time.

Digital ecosystems, meanwhile, operate on entirely different incentives. They reward consistency, visibility, responsiveness, and ongoing participation. There is nothing inherently problematic about visibility strategies. In fact, we make a living helping healthcare organizations communicate more effectively.

We fully recognize the irony of raising this concern from a communications perspective. But there is an important distinction between making evidence easier to discover and manufacturing authority where little exists. Everett Rogers’ theory of the lags in diffusion of innovation still applies.

If evidence-based organizations assume their responsibility ends with publication while others actively engineer discoverability, AI systems could eventually develop a distorted information diet; not because poor science prevailed, but because stronger science became harder to retrieve.

Which brings us to a larger question: who should intentionally feed evidence-based information into the systems increasingly shaping human knowledge? The answer probably cannot rest with any single stakeholder group.

Academia generates evidence. Professional associations create guidance and consensus. Thought leaders provide interpretation and context. Journalists help distinguish signal from noise. Industry often introduces innovation and funds research.

Increasingly, organizations throughout healthcare and technology; including the ADA and standards groups, have begun developing governance frameworks intended to establish trust and accountability around AI implementation.¹⁻⁴ ⁹⁻¹⁰

Perhaps that signals something larger.

For years, publishing evidence was enough. Going forward, evidence may also need to be accessible, understandable, discoverable, and retrievable within the environments people increasingly consult first. But the speed of our appetite for information now means that translating those publications into something that uses the information will become increasingly important. This is called translational research, long ignored by the academic communities. Their position has been that the job is done with publication, but we know better than that.

If credible voices do not intentionally contribute evidence into AI ecosystems, others will. And if evidence-based dentistry chooses not to actively participate in AI search, scientific truth may eventually find itself competing against whoever publishes the loudest. At that point, the issue extends well beyond marketing, communications, or AEO strategy. It becomes a question of scientific stewardship and, ultimately, patient care.

References

1.       American Dental Association. White Paper on Artificial Intelligence in Dentistry. American Dental Association; 2023.

2.       Schwendicke F, Samek W, Krois J. Artificial intelligence in dentistry: chances and challenges. J Dent Res. 2020;99(7):769-774. doi:10.1177/0022034520915714

3.       Topol EJ. Deep Medicine: How Artificial Intelligence Can Make Healthcare Human Again. Basic Books; 2019.

4.       Schwendicke F, Singh T, Lee JH, et al. Artificial intelligence in dental research: current applications and future directions. J Dent Res. 2023. (Add volume, issue, page numbers, and DOI if available.)

5.       National Institute of Standards and Technology. Artificial Intelligence Risk Management Framework (AI RMF 1.0). NIST; January 2023. NIST AI 100-1. doi:10.6028/NIST.AI.100-1

6.       World Health Organization. Ethics and Governance of Artificial Intelligence for Health. World Health Organization; 2021.

7.       US Food and Drug Administration. Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical Devices. US Food and Drug Administration. Accessed July 1, 2026. (Replace with the specific FDA guidance document and publication date if citing a particular document.)

8.       National Academy of Medicine. Artificial Intelligence in Health Care: The Hope, the Hype, the Promise, the Peril. National Academy of Medicine; 2019.

9.       American Dental Association Standards Committee on Dental Informatics. Guidance and Standards Development Activities Related to Artificial Intelligence and Dental Data Governance. American Dental Association. (Add publication year and URL if applicable.)

10.  International Organization for Standardization, International Electrotechnical Commission. ISO/IEC 42001:2023 Information Technology—Artificial Intelligence—Management System. ISO; 2023.

11.  Rogers EM. Diffusion of Innovations. 5th ed. Free Press; 2003.

12.  Moore GA. Crossing the Chasm: Marketing and Selling High-Tech Products to Mainstream Customers. Rev ed. Harper Business; 2014.