Benefits Leaders Do Not Have to Figure Out AI Alone

Janine Gianfredi, Chief Marketing Officer, Transcarent

If you work in healthcare and benefits, your social feed is likely full of content about how AI can improve lives, enhance employee wellness, and save money.

Sixty-six percent of HR professionals believe that AI can enhance the employee experience¹. It is an innovation that has been a catalyst for meaningful change. However, for many HR and benefits leaders, the AI conversation creates a real challenge: not just understanding what AI is, but also evaluating it, deciding how to use it, and figuring out where it can add value.

It was, unsurprisingly, a major topic of conversation at the annual Northeast Business Group on Health (NEBGH) conference, where employers and benefits leaders explored the future of healthcare and benefits.

At Transcarent, AI has been a key part of our Member experience for a few years. This year at NEBGH, there was a notable difference in the conversation. Benefits leaders are no longer asking “Do we need AI?” Instead, they are asking more in-depth questions:

• How does AI outperform the old model?

• Is this a paradigm shift or a passing trend?

• How can I use AI most effectively?

• Does AI reduce friction and improve outcomes?

I was part of the panel, AI in Benefits: What’s Working, What’s Coming, and this is exactly the discussion we dug into. From benefits navigation and care access to plan administration and strategy, we addressed the questions employers should be asking now.

Making Informed Decisions About AI

We know you are under pressure to identify the best use cases of AI as it shows up in vendor pitches, renewal conversations, and strategy discussions across the benefits ecosystem. At the same time, you have to stay focused on outcomes, responsible governance, and the employee experience. That pressure is justified: nearly 50% of HR decision-makers using AI for benefits administration reported improved utilization, along with gains in wellness, efficiency, satisfaction, and reduced leave.¹

Another shift with AI is that the decision to bring new tools and platforms into your organization is not one made in a silo. Your procurement, compliance, legal, and finance teams are weighing in on the decision, as the technology is rapidly advancing. You need a common standard for comparison. The good news is that you do not need to be an AI expert to make these decisions. You simply need to know what questions to ask.

Better AI Questions Create a Better Way Forward

At Transcarent, for years we've led the industry in responsible AI innovation by pioneering AI standards and setting precedents. We’ve been listening to the needs of employers and their people, and have created a way for them to separate meaning from the noise when evaluating AI. I'm excited to share this practical framework to help benefits buyers focus on the capabilities that will deliver real value for Members and employers.

Here are six questions every benefits leader should ask AI vendors:

1. Is the AI platform more than a chatbot? The first wave of AI launched a plethora of tools that did one thing: answer simple questions. With agentic AI, we’ve moved beyond that and now need to ask if it can take meaningful action to connect a Member to care.

2. How are privacy, security, and reliability managed? These are non-negotiables and start with learning what certifications a company has earned. Next, you need to ensure that your employees and their families can trust the platform, which means asking how sensitive data is protected, who has access, and what controls are in place.

3. What decisions can the AI make on its own? How proactively can the system act, where are its boundaries, and how well do those boundaries match an employer’s risk profile? For instance, can the platform identify the right next action and schedule it, or does it still leave the Member to do the work alone?

4. When and how is human expertise brought in? The best AI platforms will bring in clinical judgment or specialized support when needed and make a handoff without losing context. You should ask what triggers a handoff to a human and how the solution ensures a Member does not have to start over when support is more complex or clinically important.

5. What evidence is shared to prove the value? It is important to see measurable outcomes, not just technical claims, including proof that the platform improves results without creating frustration. Is there data that shows the platform can successfully resolve the majority of inquiries instantly and other meaningful figures to show it is more than just a technical claim?

6. What data truly powers the AI? Benefits leaders should understand the depth, quality, and connectedness of the data behind the AI, including claims, clinical, and provider information. Find out whether the data is powered by deep, connected data architecture and millions of interactions.

Want to know more? Get deeper insights into each question and learn how to identify overstated capabilities, distinguish legacy AI features from purpose-built AI, recognize responsibly developed solutions, and focus on what truly delivers value.

Watch Transcarent's on-demand executive briefing, The Benefits Buyers Briefing: 6 Questions to Ask on Agentic AI, hosted by Anjali Jameson, Chief Product Officer, and me. We explore the thinking behind each question and share a practical framework for evaluating AI vendors as benefits leaders make bold decisions about the future of benefit design.

It's time to look beyond a polished demo and evaluate whether an AI solution can deliver better decisions, better experiences, and better outcomes for Members and employers.

Source

1. Transcarent, “AI at Work: Trends, Impact, and What’s Next for HR and Benefits”