AI and the Youth Mental Health Crisis: We're Still in the Typewriter Years

I still remember the clatter of a typewriter. A metal frame, an inked ribbon, letters struck one by one on the page. No spellcheck, no undo button, no way to review what you'd written last week. Early computers were little more than improved typewriters - word processors that let you delete and edit without starting over. Watching those machines, few imagined they would one day connect us all, drive video calls, or fit into the palm of our hands. The transformation didn't happen overnight; it took a shift from asking what the new tool could replace, to asking what it could connect.

I think we're at a similar inflection point with AI.

Every week, the conversation swings between pessimism and hype. Will AI steal jobs? Is it a bubble? The start of something wonderful, or the end of the world as we know it? Some concern is justified - each new technology has sparked its share of fear and resistance. But what if, instead of fearing the end, we see this as a chance to solve problems that have long frustrated us: diseases that resisted cure, energy and food systems stuck in old patterns, and, closer to home, a youth mental health crisis that no one has yet managed to contain?

That last point is where my work has been focused for nearly three decades.

What AI is already doing, quietly

AI's most promising role in mental health is often overlooked because the popular image is a chatbot giving therapy sessions. That is a narrow view - and therapy bots are not therapists and never should be. Instead, AI's value lies largely in triage: providing low-intensity support for everyday stress, screening at scale, and allowing human clinicians to spend time on the cases that truly need them.

Early detection is further along than many realise. AI can now spot the subtle language patterns that signal early warning signs of depression, anxiety, or suicidal thinking, across different languages and cultures. Some systems are even beginning to match individuals to the kinds of help they need, moving quickly from flagging risk to connecting people with care. While still in early stages, these developments show a clear capability, like the first word processors proving the case for a new category of tools.

The gap AI could actually close

The real opportunity is that AI can close a long-standing gap in our system: coordination. Effective interventions exist - by therapists, counsellors, charities - but these good ideas are too often trapped in separate silos. A young person's distress may be spotted in one part of the system, while the help they need is elsewhere, out of reach or unknown. We already know many young people turn to online chatbots for emotional support rather than waiting for, or even seeking, a therapist. That tells us the formal system is often too slow, too distant, or simply not there when it matters.

AI, carefully designed, could be the connective tissue - the layer that links early warning signs to appropriate support, whether self-help, school counselling, or urgent referral. This isn't glamorous work; it's plumbing. But a system that helps services talk to each other can reach everyone, from a village school to a sprawling city - and it doesn't require anyone to run the whole show alone. That is a capability no single agency or government has managed on its own.

What has to be true for this to work

None of this happens by accident. AI systems must be designed from the start with clinical oversight, transparency about data, and, crucially, with young people involved in the build. A tool that ignores how they live and communicate won't be trusted or used. Nor should AI be expected to replace the adults who listen, guide, and care - it should make sure the right person or support is there when a young person reaches out, wherever they happen to be.

Where this leaves us

We're a little further along than the "typewriter years" of technology, but not by much. The AI tools that truly reshape mental health care are not yet fully formed. What I'm certain of is this: AI's potential lies in connecting what was always there, bringing together people and resources in a timely, joined-up way.

The technology was never the missing piece. The connection was.

← All articles