bswift has introduced a major upgrade to Emma™ Chat, enhancing its AI-powered benefits support experience with personalized, plan-specific guidance and seamless transitions to live human assistance. Powered by the company’s Emma Intelligence™ platform, the latest version is designed to help employees resolve benefits questions faster while reducing strain on service teams.
The enhanced Emma Chat uses each employer’s benefits configuration alongside employee-specific eligibility, enrollment, and—when enabled—claims data to deliver highly contextual responses. This approach allows Emma to go beyond generic answers, offering guidance grounded in real plan details and anticipating common points of confusion during moments such as open enrollment, life events, or coverage changes. Underpinning the experience is a multi-agent AI architecture that pulls together the right sources of information and routes each interaction toward the most effective path to resolution.
As a result, Emma Chat is achieving an 80% autonomous resolution rate. Importantly, bswift positions this metric as a balance between automation and empathy rather than aggressive deflection. “You’ll hear companies quote 90%+ containment, but too often that means they’ve optimized to discourage human support, even when it’s really needed,” said Eric Strom, SVP, Service Center at bswift. “We’ve optimized Emma Chat to resolve the right questions in the right channel: Our mix of self-service and human touch is by design.”
Also Read: Deel and MoonPay Team Up to Bring Stablecoin Salary Payments to Global Workforce
The latest update also deepens Emma’s ability to proactively guide employees using claims integration. By aligning claims signals with plan rules and eligibility context, Emma can flag potential issues—such as deductible progress or out-of-network costs—before they escalate, helping employees take the right next step sooner.
“Claims integration helps Emma operate like a trusted advisor—thinking two steps ahead,” said Matt Waldrup, EVP, Product at bswift. “The experience feels more human because it’s not just answering what someone typed in a box. It’s recognizing the context behind the question and proactively guiding the employee to the next best step.”
When complex or sensitive issues come up, Emma smoothly hands off conversations to live representatives. She shares full context to prevent repetition. In 2025, Emma handled over 750,000 messages in 150,000 chat sessions. This shows bswift’s effort to combine AI efficiency with human-centered support.
