Publications | Pharma & Healthcare AI Research Papers & Articles

June 2026 PharmaTools.AI Whitepaper

Featured
Designing Trustworthy AI: Why High-Stakes Systems Are Converging on Verification

A working paper arguing that trustworthy AI in high-stakes domains is an architectural problem as much as a model-capability one. It names the generation → verification shift, treats fabricated citations and confabulated rationales as structurally predictable failures of generative architectures, and uses healthcare as the stress test for what verification-first design requires.

Verification AI Healthcare AI Hallucinations
July 2026 Zenodo Preprint

Evaluation as Infrastructure: Deterministic Regression Gating for Evidence-Grounded AI Systems in Production

An experience report applying OpenGATE to four production systems, including an MHRA-registered medical device. Three of the four yielded real defects on their first run: a silent parse failure affecting roughly half of multi-claim verdicts, two name-capture failures in a de-identification engine, and a simplifier that dropped an antibiotic dose from a discharge summary. A controlled comparison against RAGAS and DeepEval finds the two frameworks disagree about what their shared “faithfulness” metric means — on outputs asserting figures absent from the source, one detects 5/6 and the other 0/6.

AI Evaluation LLM-as-Judge Regression Testing
May 2026 AI Advances Article

Your RAG Passes Every Retrieval Metric — And Still Lies to You

Recall@k confirms the right document was retrieved but says nothing about whether the answer is faithful to it. Introducing LitRAG, a ~200-line open-source pipeline with a two-stage citation check — deterministic quote localization plus LLM-as-judge support grading — that catches fabricated quotes and unsupported claims retrieval metrics are blind to.

RAG Citation Faithfulness AI Evaluation
May 2026 Towards AI Article

I Built an RSI for My RSI

Building a Recursive Self-Improvement pipeline that detects Repetitive Strain Injury — and the two-stage validation gate that keeps the self-evolving detector accurate, clinically plausible, and free from reward-hacking shortcuts.

Recursive Self-Improvement AI Safety Reward Hacking
Mar 2026 SN Compr. Clin. Med. Journal

Translation, not Interpretation: Rethinking Language Model Design for Healthcare

A conceptual perspective proposing that constraining LLMs to translational tasks — restructuring information across clinical, scientific, regulatory and patient-facing domains — may reduce ambiguity about authority and improve safety, auditability and accountability in healthcare AI.

Healthcare AI Clinical Communication AI Governance
Dec 2025 AI Advances Article

What Happens When AI Runs Out of Data?

Exploring the implications of data scarcity for AI — and why the future of machine learning depends on models that learn from the world, not just the web.

AI Training Machine Learning Experiential Learning
Nov 2025 ESMO Congress Conference

PosterLens: AI-Powered Mobile App for Decoding Scientific Posters

ePoster presentation showcasing PosterLens, a mobile application that uses AI to extract and simplify clinical trial data from conference posters.

Mobile AI Clinical Trials Conference Presentation
Sep 2025 medRxiv Preprint

Validation of an AI-powered mobile application for personalizing medical note explanations

Three-phase validation of Patiently AI: computational readability analysis, expert evaluation by 15 healthcare professionals, and a patient survey with 54 participants. Expert reviewers rated outputs 4.49/5 for accuracy and 4.53/5 for clarity.

Health Literacy Clinical Validation Patient Engagement
Aug 2025 Medium Article

When AI Can't Forget: The 7 Hidden Risks of LLMs in Healthcare

Seven critical vulnerabilities in healthcare AI systems, from model inversion attacks to poisoned training data, with actionable strategies for building robust defenses.

Healthcare AI Data Privacy LLM Security
Jul 2025 Generative AI Article

How I Won a Major Healthcare Communications Award… With just my laptop

The story of winning a Communiqué award by building Patiently AI — a practical case study on the six-month, zero-funding process as a solo developer.

Award Winner Healthcare Comms AI in Healthcare
Jul 2025 Generative AI Article

Why LLMs Love JSON

How structured JSON inputs can replace fragile LLM prompts, enabling scalable, compliant, and modular AI-generated content with real-world use cases.

LLMs Structured Output AI Engineering
May 2025 AI Advances Article

How to Break a Large Language Model

How and why LLMs fail under cognitive pressure — especially in complex, multi-step reasoning tasks. Real-world examples and adversarial prompt techniques.

AI Safety LLM Reasoning AI Evaluation
Apr 2025 Pharmaphorum Article

How to Fix Pharma's Content Duplication Problem

How AI-powered content adaptation can reduce waste and align messaging across medical and commercial teams, while maintaining regulatory safeguards.

Content Strategy AI Workflows Medical Affairs
Feb 2025 Medium Article

From Generative AI to Agentic AI: Pharma's Next Revolution

How agentic AI is taking pharmaceutical innovation beyond generative AI, creating autonomous systems that can independently research, analyze, and make decisions.

Agentic AI Pharma Innovation Autonomous Systems
Jan 2025 JNGR 5.0 Journal

A Day in the Life of an MSL Powered by AI

A revolutionary concept for MSL training through seamless integration of Generative AI, RAG, Multimodal AI, and AI Agents.

MSL Training RAG AI Agents
Jan 2025 AI Advances Article

How I Built Smarter AI Tools for Pharma: From Generative AI to Agentic AI

A technical deep dive into building advanced AI tools for pharma, including the transition from generative models to agentic systems.

AI Development Agentic AI Technical

Building AI for pharma, compliance, or patient communication?

Get in touch →

Research, articles & presentations