By Duncan McDonald with guest contributor, Edye Edens JD, MA, CIP, CCRP,  Founder and CEO, EEDEE Law Edye Edens Life Sciences Law Group

The Collision of Code and Compliance

Clinical trials have never moved faster. AI now writes sections of protocols, flags data anomalies before humans can blink, and even helps identify trial sites through predictive recruitment models.

Yet, with every digital advancement, there’s a new set of ethical and legal boundaries being redrawn.

As AI enters decision-making processes once reserved for trained professionals, the question isn’t just “Can we automate this?”  it’s “Should we?”

And sometimes, that’s the moment you don’t need another algorithm, you need a lawyer.

1. When the Data Talks Back

From eConsent systems to eCOA, data flows in real time across global networks. Every patient interaction creates a digital footprint and with it, a compliance exposure.

Edye Edens: “Many research organisations are unaware that each AI-powered workflow effectively creates a new data-controller relationship under GDPR and HIPAA. The minute your ‘bot’ interprets or transforms data, you’re potentially changing your legal responsibilities. Technology moves in seconds whereas regulation interprets in years. That’s why proactive risk and privacy assessment must precede automation.”

This is where compliance and code must shake hands early. A good AI model should be explainable, bias tested, and audit ready. The “bot” may be efficient but only a sound legal and quality framework ensures that efficiency isn’t later viewed as negligence.

2. When Automation Outpaces Accountability

Digital monitoring tools and adaptive IRT systems now make autonomous decisions about drug supply, dosing windows, and patient scheduling. But automation doesn’t absolve accountability.

If a trial participant misses a critical dose due to an algorithmic error, the responsibility isn’t shared with the server. It’s human.

Edye Edens: “Delegation in research is still delegation, whether to a coordinator or a codebase. Legally, sponsors and investigators retain ultimate oversight. Under ICH E6(R3), that duty of supervision is non-transferable. AI can support decisions, not replace them.”

As the FDA and EMA pilot AI/ML regulatory frameworks, the safest sponsors are those building “effectiveness checks” into their systems from day one, embedding human review checkpoints at every critical juncture.

3. When Bias Becomes a Breach

AI learns from data and data carries the bias of its creators. In clinical research, that bias can become an ethical fault line.

Consider recruitment algorithms that unintentionally under-represent minority populations due to historic data skews. In the age of digital ethics, that’s not just a DEI issue, it’s a research equity and compliance risk.

Edye Edens: “Bias in AI-driven enrollment can cross into research ethics territory if it systematically excludes vulnerable populations. Regulators are already signalling that algorithmic fairness will be treated as an ethical obligation, not a Corporate Social Responsibility (CSR) initiative.”

Legal frameworks are catching up, but until they do, ethical governance must lead where regulation lags.

4. When “Efficiency” Looks Like Misconduct

Clinical teams often deploy automation under the banner of “efficiency.” But efficiency without ethics is a disaster waiting to happen.

Tools that summarise medical notes, generate site reports, or draft protocol amendments must be validated under the same scrutiny as any manual process.

Edye Edens: “AI output is still your output. If an automated system fabricates or misinterprets trial data, it can trigger the same regulatory consequences as falsification under 21 CFR Part 312 or EudraLex Volume 10. The presence of a ‘bot’ doesn’t dilute the obligation of truth.”

That’s the pivot point, when efficiency starts eroding accuracy, it’s time to slow down and bring in life sciences counsel.

5. When Innovation Needs Interpretation

Innovation in life sciences doesn’t just create new tools; it creates new precedents.

AI-assisted safety monitoring, adaptive consent pathways, and predictive patient engagement platforms are redefining what’s possible but also what’s permissible.

The challenge is translating innovation into compliant action.

Edye Edens: “Our role as counsel isn’t to slow innovation; it’s to protect it. The strongest sponsors and vendors are those who involve legal and ethics experts at the whiteboard stage, not after the rollout.”

The Takeaway: AI Can Accelerate Trials, but Only Law Sustains Them

There’s a moment in every trial where you must decide: Is this a problem of scale or a problem of scope?

If it’s scale, bring in the bot. If it’s scope, the kind that redefines responsibility, privacy, or ethics, bring in the lawyer.

The future of clinical trials depends on both.

Edye Edens: “Compliance and innovation aren’t opposing forces; they’re co-pilots. When they operate together, research becomes not only faster but safer, and far more sustainable.”

References

1. ICH E6(R3) – Good Clinical Practice (Final Guideline, 2025). International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH). Available at: https://database.ich.org/sites/default/files/ICH_E6%28R3%29_Step4_FinalGuideline_2025_0106.pdf

2. U.S. Food and Drug Administration (FDA). Discussion Paper: Using Artificial Intelligence and Machine Learning in the Development of Drug and Biological Products (2023). Available at: https://www.fda.gov/media/167973/download

3. European Medicines Agency (EMA). Reflection Paper on the Use of Artificial Intelligence in the Lifecycle of Medicinal Products (2023). Summary coverage: Regulatory Affairs Professionals Society (RAPS). Available at: https://www.raps.org/news-and-articles/news-articles/2024/9/ema-adopts-reflection-paper-on-ai-ml-in-drug-devel

About the Authors

Duncan McDonald is the founder of eClinical Edge, a thought-leadership newsletter series exploring the intersection of technology, AI and clinical research transformation.

Edye Edens, JD, MA, CIP, CCRP is Founder and CEO of EEDEE Law, a boutique law firm dedicated to the life sciences industry. She specialises in research ethics, compliance, and the legal frameworks governing AI in clinical research.