Why AI Trust Comes Down to One Equation: Q&A with Parth Khanna

Over recent years, the pharmaceutical industry has generally embraced AI. Due the nature of the technology, various companies have implemented it across multiple sectors. Now, the question that everyone is asking is: where is AI having an actual impact?

In March of this year, ACTO co-founder and CEO Parth Khanna wrote a piece for Pharmaceutical Executive discussing an area where he sees AI making a major impact. According to him, digital health solutions allow companies to both gather larger amounts of data and then derive actionable insights from that data.

Khanna recently spoke with Pharmaceutical Executive, where he elaborated on this topic, while also discussing how things have evolved in the months since he initially wrote his original piece.

Pharmaceutical Executive: What AI implementations produce real business value?
Parth Khanna: The most important factor in any AI rollout is the trust the program can generate. When you study where that trust comes from — how a human being comes to trust artificial intelligence — two big components emerge consistently, both through our work with customers and through a series of executive roundtables we've had the privilege of conducting. It almost comes down to an equation: trust equals compliant plus capable.

Think about it in everyday terms. If you have an AI tool that is capable but not compliant — take ChatGPT in its bare form, where it sometimes goes off in unintended directions — you're probably not going to trust it fully. Conversely, if you have something that is highly compliant but not capable — think of the AI assistant that pops up on a website and responds to every question with "Sorry, I can't answer that" — you'll eventually conclude it's a waste of time. Trust disappears in both cases. Being compliant and being capable are both essential.

What our research and experience have led us to is this: the more context an AI system has about the role and responsibilities of the human it's designed to support, the more capable and compliant it can be simultaneously. Take a medical science liaison as an example. An AI agent that understands what an MSL does, what a scientific exchange looks like, what the parameters of that role are, and the specific ways it should and shouldn't support that person — that system will be both more capable and more compliant than one built without that context. The role knowledge is what makes the difference.

PE: How is AI changing what reps can do in the field?
Khanna: It's an exciting time to be enabling and supporting life sciences organizations and their frontline teams with AI. Just last week, for example, I was speaking with a chief commercial officer whose company is bringing a very exciting new treatment-resistant hypertension drug to market. The conversation was about how to give frontline teams every possible competitive advantage — and what that actually looks like in practice.

There's a prevailing debate in the ecosystem about the role of the sales rep. Do we even need a rep in a world where HCPs are getting information from tools like OpenEvidence or OpenAI? The answer that keeps resonating is that sales reps add the most value when they help physicians and healthcare professionals apply clinical data and evidence to their own practice. HCPs can get answers faster than ever — but what they're still looking for is a knowledgeable partner who can help them work through the implications of that information in the context of their patients. That's where AI can be immensely helpful, from pre-call planning and answering critical product questions all the way to coaching support and helping reps pull through the development feedback they've received from their managers.

This brings me back to a framework I think about a lot: building both human intelligence and empathetic agents simultaneously. Human intelligence comes through AI-powered adaptive learning — ensuring reps get trained faster and are field-ready sooner — and through AI-based role-play practice, where reps can rehearse in front of an AI avatar before they speak to a real customer. They get more repetitions in, and they can land the message with greater confidence and impact.

Empathetic agents come into play at the point of need — pre-call planning, HCP targeting, surfacing complex product information in real time. Human intelligence working alongside empathetic agents is, in our view, the winning formula for field sales teams today.

PE: How is AI being alongside compliance restraints?
Khanna: When it comes to compliance and governance, the first and most important step is building and designing agents the right way from the start — through what we call context engineering. The more context an agent has during development, the better it understands the standard operating procedures and guardrails it needs to operate within. That foundation is essential.

The second key element is building role-based agent systems. When agents are designed around a specific role, they can inherit the compliance protocols and guardrails that correspond to that role. If I'm designing an AI agent system for a sales rep, for example, I already know from the human world what a sales rep can and cannot do — and many of those compliance requirements can be built directly into the agent. Roles at the center of the design process is not just a preference for us; it's an essential architectural principle.

Once you've designed your agents in a robust, role-specific way, the next critical step is observability and testing. Rather than deploying agents into the field and hoping they give the right answers, the right approach is to examine and certify agents in a test environment before deployment — verifying that they respond correctly to a defined set of questions and scenarios. This is no different from how we train and onboard field teams: we make sure people are certified and pass a knowledge assessment before they stand in front of a customer. We should hold our agents to the same standard.

Finally, on the notion of ongoing observability: the goal is to move agentic systems from a black box to a glass box. That means continuously monitoring agent responses after deployment — and giving agents the same assessments you ran before launch at regular intervals, to catch any drift before an agent starts saying or doing something it shouldn't.

PE: What is the future of field reps and their relationship with AI?
Khanna: Based on our conversations with customers and our own thinking, we believe the role of the sales rep — and the way reps create value — is going to fundamentally evolve.

There will continue to be a very important role for humans in the field. But the job description of a sales rep is going to change in a meaningful way. The tasks that are currently done manually or in a human-only way will increasingly be augmented by compliant, empathetic agents. And that augmentation will free reps to go further and deeper into areas of work they can't fully access today.

We also think there's a world where the sales rep and MSL roles become hybridized — though the regulatory environment will need to catch up to that possibility. The fundamental driver is this: agents that perform at the periphery of human cognition allow humans to do more. That might mean taking on more complex clinical questions with healthcare professionals, spending more time building genuine relationships, or developing a deeper understanding of how a physician actually runs their practice. There are likely ways that role will evolve that we haven't even imagined yet.

What we can all agree on is that the future sales rep will be one who creates more value for physicians — capable of doing more, both cognitively with the help of agents, and with the time that gets freed up when agents handle the work that doesn't require human judgment.