What a 21-Year-Old Built Alone── ToneAdapt's Growth and the Accountability Gap in an Age of Accessible Creation

Kyan Santiago-Calling, age 21, built ToneAdapt — a tone translator that analyzes guitar and bass sounds from recorded tracks and returns settings recommendations calibrated to the gear a user actually owns. The story is interesting on its own terms. But the question it raises for pharmaceutical and medical work is different: when the cost of building falls to near zero, what does that mean for the domains where the cost of getting it wrong is not a wrong guitar tone, but a patient decision made on inaccurate information?

01One Specific Frustration

In 2024, Kyan became absorbed in 1980s heavy metal. He played guitar but could not get the sounds he wanted. Guitar tone is determined by the combination of instrument, amplifier, effects pedals, and the specific settings on each — a combination space vast enough that a beginner has no obvious starting point.

Around December 2024, he decided to solve the problem himself. He described it as a portfolio project. The framing matters: this was not a venture-backed startup. It was one person, motivated by a specific frustration, using available tools to build something that did not yet exist.

02How ToneAdapt Works

The product's value proposition is simple: you want to sound like a specific track; you get settings recommendations for the specific gear you own. The scale of the database determines how useful it actually is.

03Zero Ad Spend, 115,000 Users

ToneAdapt launched on the App Store on April 12, 2026. Within four months it had accumulated more than 115,000 users. Growth came entirely from organic posts on TikTok and Instagram, some of which reached several million views.

Guitar content performs well on short-video platforms: the sound is audible, the physical movement is visible, the result is immediate. This distribution dynamic is not limited to music. Patient communities and clinical professionals use the same platforms. A video demonstrating a symptom management technique can accumulate millions of views on the same mechanism that spread guitar tone content. The reach is real. So is the need for the information behind it to be accurate.

04AI as Construction Partner

Kyan used AI tools throughout — code generation, logic design, interface iteration. He did not bring an unusual professional background. The technical cost of going from idea to shipped product has dropped substantially. That creates a genuine opportunity for people who identify real problems and have the will to address them.

It also raises a question that does not go away. When the cost of building falls, the cost of distributing something inadequate falls with it. In domains where the consequences are aesthetic — the wrong guitar tone — the stakes are contained. In domains where they are clinical, they are not.

05What Can Be Built vs. What Must Be Verified

Placing ToneAdapt's example alongside regulated healthcare contexts makes the structural gap visible.

06When the Domain Is Medical

The same technical pathway that produced ToneAdapt is available to anyone building a symptom tracker, a medication reminder app, or a clinical decision support tool. The ability to build and reach users at scale is no longer restricted to organizations with engineering teams and distribution budgets.

In Japan, software meeting the definition of a medical device program requires notification or approval from the Ministry of Health, Labour and Welfare under the Pharmaceuticals and Medical Devices Act. Clinical evidence requirements, manufacturing and distribution authorization, audit records — these exist because the cost of a false or inaccurate output is not a wrong guitar tone, but a patient decision made on unreliable information. The regulatory structure does not thin out because development costs have fallen. As the number of people who can build increases, the gap between what can be built and what can responsibly be distributed becomes more visible, not less.

07Keeping the Question

Kyan Santiago-Calling started from one specific frustration: he could not reproduce the guitar sounds he wanted. 115,000 people turned out to share a version of that same frustration. None of that was predictable. What he did was identify the problem precisely, build something that addressed it, and reach the people who needed it.

That sequence — identify, build, reach — is available to people working in pharmaceutical and medical contexts too. AI has changed the cost of the middle step. It has not changed the importance of the first: knowing exactly what problem matters and for whom. And in regulated medicine, it has not changed what must come after the third: verification, accountability, and continued scrutiny of whether what was built actually does what it claims. Lower barriers to building make those post-distribution obligations more visible, not easier to skip. Holding that question, in each day's work, is what コラム means when it is not passive acceptance but active care.

ToneAdapt's story is straightforward in the context it belongs to: passion combined with a precisely identified problem, AI-assisted construction, and organic distribution can produce real reach without institutional backing. In music, that is a clean account of what the current tools make possible.

In regulated domains, lower barriers to building do not lower the accountability required once something reaches users. The gap between what can be built and what can responsibly be distributed becomes more prominent as building becomes easier — not because the rules changed, but because more people are now on the building side of that line.

For pharmaceutical and medical professionals, the relevant questions are the ones that do not get automated away: what problem genuinely matters, for whom, and what verification does the answer require before it reaches the people who will act on it. Those questions are the work.