Your professional development catalog has a leadership certificate track, a library of self-paced courses, probably an AI coaching app bolted on last year. Your best charge nurse wants to move into a supervisor role but can't find thirty uninterrupted minutes to work through the leadership module before her next shift. Your highest-potential engineer keeps getting passed over for lead roles because nobody's actually helped him practice the hard conversations that come with the title. Your top account manager is thinking about leaving, not over pay, but because she feels like nobody has invested in where she goes next. None of the three has opened the AI coaching app once this quarter.

Human coaching vs AI development tools is landing on L&D desks everywhere right now, and the stakes are bigger than which line survives the next training budget review.

What Is Professional Development Actually For?

Most organizations answer this by pointing at a course catalog. Vineet Rajan, Forte's Co-founder and CEO, starts somewhere else: with the person the catalog is supposedly for.

"Professional development is to develop a person within a profession," Rajan says. "It's not there to develop an algorithm, a machine, or a robot. If you only focus on just the profession piece, you lose the actual vehicle delivering the services and goods that all of your customers, clients, and stakeholders want and need from you."

The technical side of a job, a new certification, a new procedure, a new compliance module, matters. But Rajan points to what actually separates a competent professional from an exceptional one: "The things that often separate a person who's really competent in their job, oftentimes it isn't a technical piece, it's an interpersonal piece. It's the dynamic thinking, the communication, the empathetic ability to see and hear another person." Ask any superintendent what actually drives student outcomes, he says, and the honest answer is staff, and what makes one math teacher better than another isn't the math. It's their ability to bring it alive for the kids in front of them.

That distinction is getting sharper, not softer, as AI tools get better at the technical bits. "As friction falls by using tools like artificial intelligence, we've lost the muscles for empathy, for communication, for creativity, for curiosity, for patience," Rajan says. The parts of a job that are hardest to automate are exactly the parts professional development is supposed to build, and they're not built by a webinar.

What Do AI Coaching Apps Actually Promise a Development Program?

The market backing these tools is real money: mental health and coaching apps, the same category increasingly bundled into professional development platforms, are valued around $7.7 billion in 2025 and projected to reach $26.5 billion by 2033, a 16.65% annual growth rate (Grand View Research, 2025). The pitch to L&D leaders is consistent: 24/7 access, no scheduling friction, infinitely scalable across a workforce of thousands, priced far below a coaching program. Those are real advantages. A night-shift employee can open an app at 2am. She can't book a coach at 2am. Access matters, and dismissing it isn't honest.

But the market has outrun the evidence on whether access builds capability. A 2025 meta-analysis of 15 randomized controlled trials found AI-delivered apps produce small to moderate short-term effects, with long-term benefits largely not sustained; effects on sustained focus and positive affect were often not statistically significant. Common Sense Media's Youth AI Safety Institute, working with Stanford Medicine's Brainstorm Lab, ran more than 3,100 exchanges through five AI apps in this category and found the evidence for lasting effectiveness thin (Common Sense Media, 2026).

A NAMI/Ipsos poll of 2,046 adults found only 12% said they'd be likely to turn to an AI chatbot for personal coaching or support. Just 1% already did (NAMI/Ipsos, 2025).

None of that makes AI tools worthless in a development program. It means access and actual capability-building are two different things, and the market keeps conflating them.

Where the Evidence on AI Development Tools Gets Thin

Most AI app outcome studies measure gains at four to eight weeks with no control group. There's no follow-up at six months, no measure of whether someone still opens the app by week five, no account of what happens once the content runs out and nothing in the person's actual work has changed.

Researchers are still trying to define a "digital therapeutic alliance," whether a person can build a working bond with an automated system strong enough to produce lasting change. A 2025 integrative review in JMIR Mental Health found that while digital tools show promise for access and short-term engagement, the literature reviewed contained no longitudinal studies at all (JMIR Mental Health, 2025). We don't yet know whether the bond someone feels with an app in week two is still doing anything for their actual leadership growth by month six.

That gap isn't a flaw in the technology. It's a description of what a real coaching relationship does that a prompt sequence can't.

What a Human Coach Builds That an App Can't

Three researchers have spent careers answering a version of this question, and they land in the same place.

John Norcross, who led the joint APA task force on evidence-based helping relationships, found across decades of meta-analysis that the relationship between a helper and a person working toward change is demonstrably effective, independent of technique. Nine relationship elements, including alliance and empathy, were rated demonstrably effective (Norcross, APA Divisions 12 and 29, 2019).

Bruce Wampold, emeritus professor at the University of Wisconsin-Madison, built the contextual model of psychotherapy around a similar case: outcomes depend on the real relationship, the person's belief in the process, and consistent, well-explained ways of working through the problem. His meta-analytic work found a helper's interpersonal style and the strength of the working alliance often outweigh the specific method used (Wampold, World Psychiatry, 2015). Who is coaching you matters more than which framework they're running.

Lori Gottlieb, psychotherapist and author of Maybe You Should Talk to Someone, puts it plainly: transformation comes not through methods or techniques, but through the act of being heard.

A 2025 retrospective study of 201 telepsychiatry patients found working alliance independently predicted clinically significant improvement, even with every session delivered virtually (Talkiatry study, 2025). The mechanism generalizes: the medium was virtual, but the human relationship still did the work. It's the same reason a leadership coach who remembers your last three hard conversations builds more capability in you than a fresh chat window ever will.

Rajan describes what that consistency actually builds in practice: "Those reps they've been doing in composition, in endurance, in capacity, they use at work. They sell better, they teach better, they serve better, because they've been training their mental fitness and they've become more human in the process." An app can't notice you went quiet in a meeting in a way that's different from last week. It can't say, "You just handled that the way you said you wanted to three sessions ago." Whether someone feels invested in is the variable that predicts whether the growth sticks.

AI App vs. Human Coach: what each actually builds

Why Forte Keeps a Real Coach on the Other End

Forte is a Mental Fitness platform, not a course library and not a wellness app. It's a defined discipline built on four Pillars: Composition, Capacity, Flexibility, and Endurance. Composition is who someone is, their identity, values, and how they talk to themselves walking into a hard conversation, the exact thing most leadership development programs try and fail to teach from a slide deck. Capacity is bandwidth: sleep, physical health, how much someone can hold before they start dropping things. Flexibility is how fast someone recovers when a plan falls apart. Endurance is whether they can sustain growth across months and years, not just for a week after an offsite.

Every one of those Pillars needs someone tracking it with a person over time. An app can prompt someone to log a reflection. It can't ask why they've been logging it the same shallow way for two weeks or what that's really about. A Coach can, and remembers that last month someone said they were fine delegating, and this month the way they said it changed.

Rajan's argument for why that matters isn't just for the person, it's structural: a person doesn't stop being a person at 5pm. "What you experience from five p.m. to nine a.m. affects who you are from nine to five, and what you hear from nine to five affects who you are from five to nine," he says. "We have a holistic approach to what it means to be a human, and we don't expect you to separate those two worlds. In fact, those two worlds are one world."

What L&D Leaders Are Actually Buying When They Choose Coaching Over Access

Buy an AI coaching app for a 2,000-person workforce, and you're buying reach: a low-friction, always-available first touch that might catch someone thinking through a hard decision at 11pm. That's real, and worth keeping in the mix.

What that app can't sell a development program is what makes the 11pm moment matter six months later: the point where your head of operations finally admits to someone he's been improvising his management style since he got promoted, or the conversation where your rising engineer says out loud, for the first time, what's actually stopping her from applying for the lead role.

A 2025 real-world study of 266 working adults using Modern Health found that after an average of 2.5 coaching sessions, participants improved on distress tolerance, self-compassion, and mindfulness, with measurable gains that held over time (Modern Health, Administration and Policy in Mental Health, 2025). Two or three sessions with a real coach, not two or three months of daily app prompts.

Rajan makes the budget case directly, and it depends on how the line item gets framed: "If you frame this as an EAP or therapy, people think about it as a utilization game. Frame it as professional development, and it costs about what you'd spend on a meal a month, effectively less than the cost of a single executive coaching session. That's pretty wild." The math doesn't change. What changes is whether the buyer sees a benefit nobody uses or an investment in the people already carrying the organization.

L&D leaders evaluating development tools right now aren't choosing between access and quality. They're choosing whether access alone builds capability, or whether they want what actually produces the growth the program promised.

Rajan is direct about how he'd want that choice tested: "How do you measure if it's doing what you've asked them to do? Is a keynote and webinars throughout the year going to develop your profession in the way you're hoping? How do you know it's working?"