← Back to episodes

Episode 207 ·

Part 2: Artificial Intelligence (AI) & Fair Market Value (FMV) under Stark Law & Anti-Kickback Statute (AKS): A Discussion with Chris David, Managing Partner, HealthValue Group

Send us Fan Mail Artificial Intelligence (AI) needs a little pushback. In Part 2 of this 2-part episode, Captain Integrity Bob Wade continues his talk on the intersection of AI, Fair Market Value (FMV), the Stark Law, and the Anti-Kickback Statute (AKS) with HealthValue Group Managing Partner Chris David. Hear why we need to be human to make AI helpful, how to protect your nonproprietary data AI can’t replicate, why survey data & valuation models are a commodity now, why AI won’t replace you as healthcare consultants, and an exercise involving a letter generated by Copilot. Learn more at CaptainIntegrity.com

  • Fair Market Value
  • Anti-Kickback Statute
  • Artificial Intelligence

Listen to the episode

Audio

Ready to play. Audio loads only after you press Play.

0:000:00

Prefer Buzzsprout? Listen on Buzzsprout.

Companion article

Part 2: Artificial Intelligence (AI) & Fair Market Value (FMV) under Stark Law & Anti-Kickback Statute (AKS): A Discussion with Chris David, Managing Partner, HealthValue Group

Episode Date: March 3, 2026

In this episode of Stark Integrity, Bob Wade (Captain Integrity) continues the discussion with Chris David, Managing Partner at HealthValue Group:

A deeper dive into how artificial intelligence (AI) intersects with fair market value (FMV) and compliance under the Stark Law and Anti-Kickback Statute (AKS).

This episode is Part 2 of a two-part series, focusing on practical application, real-world examples, and the limitations of AI in highly regulated environments.

The central message:

AI is a powerful tool—but without human oversight and critical thinking, it can fall short of compliance expectations.

Moving from Theory to Practice

While Part 1 focused on concepts, Part 2 shifts to:

How AI performs in real-world FMV analysis scenarios.

The discussion explores:

  • AI-generated valuation analyses
  • How outputs compare to traditional methods
  • Where gaps and risks emerge

The takeaway:

Seeing AI in action reveals both its strengths and its weaknesses.

Evaluating AI-Generated FMV Analysis

A key element of the episode is:

Reviewing an AI-generated FMV analysis and assessing its defensibility.

This exercise highlights:

  • Gaps in reasoning
  • Missing context
  • Lack of nuanced understanding

The key point:

AI can generate content—but not always a compliant or supportable analysis.

The Limits of AI in FMV

As emphasized by Chris David, AI has clear limitations when applied to FMV:

  • It may rely too heavily on generalized data
  • It may not account for specific facts and circumstances
  • It may miss key regulatory nuances

The takeaway:

FMV requires judgment—and judgment cannot be automated.

The Importance of Being Human

One of the most important insights from the episode is:

Human input is essential to make AI useful.

Professionals must:

  • Interpret AI outputs
  • Adjust for context
  • Apply experience and judgment

The key point:

AI works best when guided by human expertise—not when it replaces it.

Data Commoditization and Its Impact

The discussion also explores how:

Data and valuation inputs are becoming more commoditized.

With AI:

  • Market data is widely accessible
  • Benchmarking is easier to obtain
  • Tools can generate similar outputs across users

The takeaway:

Access to data alone is not a differentiator—how it is used is what matters.

Protecting Value Beyond AI

Another important concept discussed is:

Protecting nonproprietary insights and professional value.

As AI becomes more capable, organizations and professionals must:

  • Focus on interpretation and strategy
  • Provide context-driven insights
  • Demonstrate independent judgment

The key point:

The true value lies in expertise—not just information.

Why AI Won’t Replace Experts

The episode makes it clear:

AI will not replace healthcare valuation professionals or compliance experts.

Instead:

  • AI can assist with efficiency
  • AI can support data analysis
  • AI can enhance workflows

But:

  • It cannot defend decisions
  • It cannot testify or explain reasoning
  • It cannot fully understand regulatory expectations

The takeaway:

AI is a tool—not a replacement for expertise.

Compliance Risks of Overreliance

The episode reinforces the risk of:

Overreliance on AI-generated outputs.

This can lead to:

  • Insufficient documentation
  • Unsupported conclusions
  • Increased regulatory exposure

The key point:

Using AI without validation can create more risk than benefit.

Practical Considerations

To use AI effectively in FMV and compliance, organizations should:

  • Review and validate all AI-generated outputs
  • Apply independent judgment
  • Maintain strong documentation standards
  • Use AI as a supplement—not a substitute

Because:

Defensibility requires more than automation—it requires understanding.

Key Takeaways

  • AI must be tested against real-world compliance scenarios
  • AI-generated FMV analyses may lack necessary depth and context
  • Human judgment is essential to ensure defensibility
  • Data is becoming commoditized, but expertise remains valuable
  • AI enhances—but does not replace—professionals
  • Overreliance on AI introduces compliance risk
  • Validation and documentation are critical

Final Thoughts

Part 2 completes the discussion with a clear and practical message:

AI is only as effective as the humans who use it.

With insights from Chris David, it becomes evident that:

  • AI can improve efficiency
  • AI can support analysis
  • AI can streamline processes

But ultimately:

Compliance decisions require human judgment, experience, and accountability.

Because in healthcare compliance:

Technology can assist—but responsibility cannot be delegated.

Click here to listen to this Stark Integrity Podcast Episode:
https://podcasts.apple.com/us/podcast/part-2-artificial-intelligence-ai-fair-market-value/id1588939373?i=1000752989579&l=fr-FR