E208 : The future of enterprise AI: agents, automation, and trust

In this episode of AI For Pharma Growth, Dr Andree Bates speaks with Dr. Kurt Rohloff, co-founder and CEO of Duality Technologies, about how privacy-enhancing technologies could unlock data collaboration without forcing organisations to give up control of sensitive information. Kurt explains how technologies such as fully homomorphic encryption and federated learning allow organisations to analyse encrypted or distributed data without exposing the underlying patient information. For pharma and medical research, that creates opportunities to reuse existing data, collaborate across organisations and geographies, and access populations that have historically been difficult to include. The conversation explores why data security should not be viewed only as a compliance cost. Secure collaboration can also become a value creator, helping organisations run better science, reduce administrative overhead and generate evidence that would otherwise be too expensive or difficult to obtain. They also discuss rare diseases, underserved communities, cross-border research and the role regulators and standards bodies will play in enabling wider adoption. Kurt argues that the technology is often further ahead than organisational acceptance, making policy conversations just as important as technical development. Topics Covered Privacy-enhancing technologies in pharma

In this episode of AI For Pharma Growth, Dr Andree Bates speaks with Clara Bernardes, co-founder and CSO of Biorce, about how AI can accelerate clinical development without sacrificing compliance or scientific rigour. Biorce was born from a deeply personal experience. After Clara's father-in-law received a terminal melanoma diagnosis, access to a clinical trial through compassionate use extended his life from an expected three months to more than a year. That experience shaped Biorce's mission: reduce the time it takes promising treatments to reach patients. Clara explains why clinical trials are still slowed by protocol amendments, patient identification problems, poorly designed criteria and human-led errors that could often be caught much earlier. The conversation explores how Biorce uses AI and almost one million annotated protocols to support protocol development, learn from previous trial amendments and provide clear rationale for recommendations. Clara also discusses why transparency, AI ethics and keeping humans in the loop are essential in regulated clinical development. The key message is that speed and compliance do not have to be opposing forces. The goal is not to replace scientists or investigators, but to give them better information faster while preserving human judgement. Topics Covered Why clinical trials start late

In this solo episode of AI For Pharma Growth, Dr Andree Bates explores one of the biggest blind spots in pharma AI: the failures that can survive a normal review process even when the output sounds polished, plausible and well grounded. Dr Andree explains why better models and better retrieval do not remove the need for human judgement. In fact, as AI systems improve, people often scrutinise them less. Fluent prose can create a false sense of reliability, while automation bias and anchoring make reviewers less likely to challenge the underlying frame. The episode introduces three broad families of failure: invented, distorted and misdirected. These include fabricated or misattributed citations, silent interpolation, dropped qualifiers, hedge-to-claim escalation, manufactured consensus, stale guidance and planted instructions. Many of these are harder to catch because the output reads better than the truth. Dr Andree also shares practical checks teams can use, including tracing claims back to source, checking what is unsupported, looking for missing qualifiers, regenerating outputs for stability and deliberately testing the opposite case. The key message is that prompting is only half the skill. In a regulated industry, teams also need to be trained to judge what comes back. Topics Covered Why better AI can lead to less scrutiny

In this episode of AI For Pharma Growth, Dr Andree Bates speaks with cancer biologist and health educator Rebecca Maff about how AI could help shift healthcare from treating disease to preventing it. Rebecca’s interest in prevention began after seeing women in her own family diagnosed with late-stage cancers and serious autoimmune conditions. Her subsequent work has included machine learning models designed to identify patients at higher risk of cancers and encourage overdue screening before disease progresses. The conversation explores what prevention really means in practice, from early biomarkers and personalised risk modelling to screening, lifestyle, metabolic health and behavioural change. Rebecca explains how AI can help analyse electronic health records and large patient populations to identify signals that would be difficult for humans to find manually. They also discuss the commercial opportunity for pharma. If late-stage disease becomes more preventable, the industry may need to think beyond treating established illness and consider new models built around earlier detection, intervention and maintaining health. Trust remains a major challenge. Rebecca argues that patients and clinicians need to see credible, successful AI implementations before confidence grows, particularly when systems are using highly personal health, genomic and behavioural data. Topics Covered Moving healthcare upstream towards prevention

In this episode of AI For Pharma Growth, Dr Andree Bates speaks with Raviv Pryluk, co-founder and CEO of PhaseV, about why so many clinical trials still fail and where AI can genuinely improve the odds. Raviv argues that failure is often not because the drug itself is wrong. Trials can fail because of the wrong patient population, indication, dose, design, sites, monitoring or interpretation of the data. PhaseV uses causal machine learning, adaptive trial design and large-scale simulation to help sponsors make better decisions across those areas. The conversation explores why explainability and validation matter in clinical development. Raviv explains that a 95% prediction is not enough on its own. Sponsors and regulators need to understand why a recommendation is being made, which evidence supports it, and whether the result is statistically and clinically defensible. They also discuss using existing trial data to identify responder subgroups, stress-testing designs before patients are enrolled, adapting trials mid-flight and connecting protocol design more closely with clinical operations. Topics Covered Why clinical trials still fail