FVAI Vol #23

AI just outperformed humans on emotional intelligence tests.

81% vs. 56%.

June 2025 may be remembered as more than just a milestone—it feels like a turning point in how we relate to machines, and how we understand ourselves.

As someone who’s been deep in the weeds building AI voice technology, I’ve seen this shift coming. But even so—it’s jarring. Machines that once struggled to distinguish sarcasm from sincerity can now detect emotional nuance better than the average person. They can identify stress in a voice before a human ear would catch it. They respond with precision-tuned empathy—always calm, always attentive, never distracted.

They’re even creating new ways to measure emotional intelligence—tools and metrics we hadn’t thought to use ourselves.

But here’s the thing: they still can’t feel. At least not yet!

That matters. Maybe now more than ever.

Because while machines can mirror our emotions with near-perfection, what they’re reflecting is still us—our laughter, our grief, our hesitation, our joy. An algorithm might know when you’re heartbroken, but it doesn’t know what heartbreak feels like. And maybe that’s where our humanity hides—in the messy, contradictory, hard-to-define spaces that resist quantification.

So I’ve been asking myself: in a world where machines can out-empathize us on paper… what does it really mean to be human?

Is it the unpredictability? The paradoxes? The capacity to feel two things at once?

That last one hits home for me. When my daughter was little, she had a moment of confusion—she was feeling both mad and happy at the same time. She couldn’t make sense of it. And I remember helping her walk through that unfamiliar terrain: the realization that we can hold multiple emotions at once. That it’s not a glitch—it’s a feature.

As I’ve grown older, I’ve come to understand that this “gray space”—this realm of non-binary emotion—is where we live most of our lives. Somewhere between joy and sorrow, confidence and doubt, love and fear. Feeling both at once or quickly vacillating between the two.

This isn’t a lament. It’s a call to curiosity. A reminder that even as we build smarter, more intuitive machines, the soul of the interaction still begins—and ends—with us. Not because we’re perfect. But because we’re beautifully imperfect. We live in the gray. We are the gray. And we don’t need everything to be either a 0 or a 1.

So here’s my challenge to you (and to myself): let’s not outsource the practice of emotional intelligence just because machines are getting better at performing it. Let’s double down on our own capacity to listen, to understand, to show up with presence and care.

Because the future won’t just be built on intelligence—it’ll be shaped by empathy, vulnerability, and meaning.

And those—at least for now—are still beautifully, stubbornly human.

Voice isn’t just sound—it’s soul.

When we lose a loved one’s voice, we lose more than their words. We lose the way they said things—the warmth, the humor, the pride, the quirks. A part of their essence fades with the silence.

But something extraordinary just changed.

Hume.ai’s new emotional voice interface, EVI 3, can recognize 53 distinct human emotions.

For family storytelling, this is a game-changer.

Think of your dad’s voice lighting up as he talks about his first job. The flow and contour of his voice matching the feelings of the moment.

Or your mom’s laugh as she tells a favorite childhood story.

Traditional recordings preserve content.

EVI 3 preserves meaning by expressing implicit tone.

It adapts, it learns, and it captures how someone expressed themselves—not just what they said.

This isn’t a futuristic fantasy.
It’s real. And it’s here.

At Family Voices AI, we’re diving deep into this breakthrough—exploring how it can help families preserve legacies in a way that truly honors the people behind the stories.

Because when it comes to the voices we love…some things are too precious to lose.

Family Voices AI Beta Launch

It’s official. We are in BETA! Want to help us out? Please sign up for FREE HERE.