How a People Leader's AI Tool Boosted Her Benefits Confidence
The second in a series on real-world AI use cases in HR
The benefits landscape is becoming increasingly technology-driven. While traditional offerings such as health (88%), retirement savings and planning (82%), and leave (82%) remain among employees’ highest priorities, SHRM’s 2026 Employee Benefits Survey shows that newer, technology-enabled offerings are gaining ground. AI tool subscriptions saw the largest percentage increase of any benefit tracked in the survey, jumping from 16% of organizations offering them in 2025 to 33% in 2026.
That shift reflects a broader evolution in total rewards, one in which employers are not only offering new technology as a benefit but also exploring how AI can help them manage benefits more efficiently and tailor them to employees’ needs.
Across the 16 HRX dimensions, from recruitment and compliance to talent management and leadership development, AI is being used to augment the HR function. To demonstrate how everyday HR professionals are doing this, we reached out to members to share their experiences. This is the second in a new article series of real-world AI use cases in HR, this time focused on total rewards.
This article is an example of the Benefits Recommendation Tools for Employees use case. Discover 137 other AI use cases by visiting SHRM’s AI Field Manual for Employers: 138 Use Cases for AI at Work.
A Medicare Tech Company Finds an Unexpected AI Ally
Megan Peterson, SHRM-CP, leads people operations at Spark, a company that builds technology and infrastructure for Medicare sales professionals. As the sole point of contact for benefits, systems, leave, and payroll for the organization's roughly 100 employees, Peterson wears a lot of hats. Earlier this summer, Spark came off a professional employer organization (PEO) arrangement and, for the first time, got to choose its own benefits vendors.
Among those choices was Lively, an AI platform for administering health savings accounts (HSAs) and flexible spending accounts (FSAs). Peterson said the selection came through Spark's benefits broker, based on other clients' positive experience with the platform's interface and ease of use. What she didn't expect was how much Lively's built-in AI chat feature would end up doing for her team.
"Great AI support was not something that we even thought to ask about, and so one of the greatest surprises of working with Lively has been their AI feature," Peterson said. "Now going forward, I know that's something that I will be asking about, because it's been so awesome."
The AI Use Case: Benefits Recommendation Tools for Employees
The problem Peterson needed to solve was a familiar one for HR teams migrating benefits providers: Consolidation. HSAs are employee-owned, and many of Spark's HSA-eligible employees arrived with multiple accounts scattered across prior employers and plans. As Peterson put it, employees tend to be as protective of their HSA balances as they are of a 401(k) — it's their money, and they want to know exactly what's happening to it.
With a lean HR team, Peterson anticipated a flood of individual rollover questions she'd have to field herself. Instead, Lively's AI chat handled the bulk of that work directly with employees. Here's how it played out, using Peterson's own rollover as an example: An employee tells the chat what they want to do, the tool asks two or three clarifying questions — the old account number, phone number, or email tied to the prior HSA — and then produces a pre-populated form ready to send to the previous provider.
"Within minutes it was handled,” Peterson said. That reliability let Peterson redirect her own employees. Rather than fielding rollover questions herself, she was able to tell staff to take those questions straight to Lively's chat instead.
From there, Peterson extended her use of the tool beyond one-off employee questions. She used the same chat feature to help draft internal benefits instructions and enrollment materials — content she says would otherwise have taken her considerable time to write from scratch. The tool also tailors its guidance to account for variables like whether an employee's HSA funds are invested, producing different paperwork depending on the scenario.
On the administrative side, Peterson also uses Lively's AI features to pull participation data and enrollment status. For example, she asks the chat to summarize enrollment progress across a recent batch of new hires, the same way she'd ask it to walk an employee through opening an account.
Because Spark's workforce is distributed across time zones, the chat's around-the-clock availability was a meaningful part of the value. Employees could get answers on their own schedule rather than waiting on Peterson to be online.
"It was great that they could reach out and handle this whenever they wanted to, again, without me having to bottleneck and sort of guide them throughout everything they needed to do," she said.
What Worked
The clearest win was accuracy paired with speed, right out of the gate. Peterson said the tool got her own rollover request right on the first try, something she hadn't expected from a benefits chatbot.
"[The experience] was so customized and efficient — the first time I asked it, it exceeded expectations," she said.
That accuracy mattered because, as Peterson noted, credibility is central to her role. With around 100 employees relying on her guidance, she was initially cautious about leaning on AI-generated instructions. Before rolling out any chat-assisted materials to staff, she sent them to Lively's implementation team for a double check. They confirmed the guidance was correct.
The tool also handles benefits usage and forecasting questions, even though Spark's HSA program is currently employee-funded rather than employer-funded. Peterson tested prompts around participation rates and average per-employee contributions — the kind of data an employer contributing to HSAs would need to project costs.
Beyond the mechanics, Peterson pointed to a softer benefit: Employee confidence. Lively's chat feature was one piece of a broader benefits and HR systems overhaul at Spark, and Peterson believes the ease it created contributed to employees' trust in the new vendors and in the HR function's decisions overall.
"I think folks were appreciative that Lively made it easy, and that they didn't have to do a lot of paperwork on their end," she said.
What Didn't Work
Unlike some AI rollouts, Peterson doesn't have a specific failure story to point to yet — Spark is only a few months into using Lively. Any caution she’s had — testing the tool's output with Lively's implementation team before trusting it with employees — reflects an expectation that any AI tool, however good the first impression, needs to be verified rather than taken on faith.
One open question Peterson flagged is around data scope. Lively's chat handles HSA- and FSA-specific requests and, to her knowledge, isn't integrated with Spark's performance platform or HRIS, so it shouldn't have access to more sensitive employee data. But she acknowledged the tool does collect sensitive financial information as part of its normal function, such as prior HSA account numbers — meaning HR teams considering a similar tool should understand exactly what data a vendor's AI feature can access, and where the boundaries are, before rolling it out broadly.
Human Oversight and Governance Remain Paramount
Peterson's approach to trusting the tool followed a clear sequence: Test it herself, verify the output with experts, then extend it to employees. Before pointing staff toward Lively's chat for benefits materials, she ran the AI-drafted instructions past Lively's implementation team to confirm accuracy. Only after that confirmation did she start routing employee questions to the chat directly.
She's also candid that the tool has a ceiling. Some employees, she said, will always prefer a one-to-one conversation with a person over trusting a bot, and HR teams need to be ready to step in for those cases rather than assuming the tool covers everyone equally well.
"Don't fully turn over all of your operations to the bot," Peterson said. "Make sure that you're testing and that you understand it yourself, because there will be times when someone is going to value that one-to-one connection."
On data sensitivity specifically, Peterson said Spark is relying on Lively as a trusted vendor to handle financial information such as bank account numbers according to proper protocols — a reminder that adopting a vendor's AI feature also means extending trust to how that vendor secures the data flowing through it.
Advice for HR Teams
Peterson's guidance for HR professionals considering an AI-powered benefits tool:
Test the AI tool on yourself first. Before pointing employees to an AI chat feature, Peterson ran her own HSA rollover through it to see how it performed.
Verify with experts before wide release. Peterson sent AI-drafted benefits materials to Lively's implementation team for a factual check before sharing them with staff.
Keep the human option open. Some employees will always want a person, not a bot — be ready to step in rather than assuming the tool works for everyone.
Understand what data the tool can access. Know whether an AI feature is integrated with sensitive systems like your HRIS, and where its boundaries are.
Don't discount vendor AI features when evaluating tools. Peterson said AI support wasn't even a question she thought to ask when selecting Lively — it's now on her list for every future vendor evaluation.
Looking back on her own path to trusting AI tools, Peterson said she was "cautiously interested" at first. She was unsure how a technology built for efficiency would fit into a function as relationship-driven as HR. That's shifted as she's leaned into the operational side of her role.
"You get to focus on the value-adds and the really meaty stuff, and have AI handle your admin and your smaller stuff," she said.
For Peterson, the underlying goal hasn't changed even as her comfort with AI has grown: Give employees fast, accurate answers about their own money without losing the personal touch that often required for benefits questions.
Was this resource helpful?