For decades, the photography industry has lived with a quiet but consequential problem, cameras that were never built to see everyone equally.
The original colour-calibration cards that shaped photographic film in the twentieth century were built around a narrow set of reference tones, largely lighter ones.
Everything that followed, from chemical film to digital sensors and the AI models that now process our photos, inherited that same narrow starting point.
The result has been consistent and largely unremarked upon: darker skin tones rendered with flattened detail, muddy shadows, and colour casts that bear no resemblance to what the eye actually sees.
Wood grain, dark chocolate, and darker human skin have all suffered the same fate, not because of a specific design decision, but because of a bias baked so deep into the tools that most people never think to question it.
They simply assume their photos “don’t turn out right” in certain lighting, without realising the camera was never calibrated to represent them accurately.
This is the problem TECNO’s Universal Tone system is attempting to solve, and the approach is worth understanding beyond the marketing language.
Rather than accepting the industry-standard 24-patch colour reference card that has underpinned most of photographic history, TECNO built its own, expanding it to 372 distinct skin-tone patches.
That is not a cosmetic upgrade, it is a fundamentally larger dataset built specifically to capture the chromatic range of human skin across markets that mainstream camera calibration never prioritised: Sub-Saharan Africa, South and Southeast Asia, the Middle East, and Eastern Europe.
For a brand whose core markets are Africa and India, this is not a side project. It is a direct response to what its own customers have been saying for years. The credibility of the approach lies in the method, not just the numbers.
TECNO paired its expanded colour reference with a Multi-Skin Tone Colour Restoration Engine, a system designed to correct the two failure modes that have consistently plagued darker skin tones: over-darkening in low light, and unnatural reddish or ashy colour casts under artificial lighting.
Alongside this sits a local tuning engine that accounts for regional lighting conditions, climate, and colour temperature preferences that differ from one market to another.
A portrait taken in Nairobi’s midday sun and one taken under Lagos evening light are not the same imaging challenge, and treating them as identical is precisely how bias creeps back in even after a company claims to have addressed it.
The deeper issue, as TECNO appears to understand, was never really about hardware. It is about data.
When an AI model is trained overwhelmingly on lighter-skinned faces in well-lit, evenly toned conditions, it does not just perform worse on darker skin, it does not know what accurate even looks like for that skin tone.
It fills the gap with assumptions borrowed from the majority of its training data.
That is how cameras end up boosting exposure until dark skin looks grey, or applying skin-smoothing algorithms that erase texture and depth rather than preserve it.
Building a dataset intentionally weighted toward historically excluded tones and testing it across hundreds of real-world lighting scenarios is a structural fix, not a filter applied on top of the same broken foundation.
Representation in imaging has never been a luxury feature or a diversity talking point.
For the better part of a century, an entire category of people were told, implicitly, through blown-out shadows and muddy portraits, that the technology was simply not built with them in mind.
Fixing that requires rebuilding the dataset from the ground up, the harder, less glamorous work that TECNO appears to be doing.
