Business June 17, 2025

“We need an army of Elliots” – why it’s bonkers we’re not using facial age estimation to sell alcohol

profile picture Yoti 3 min read
An image of a woman trying to buy a bottle of alcohol at a supermarket self-checkout terminal.

Let’s just get this out there: humans are not great at guessing ages. Don’t just take our word for it. Studies have proven this to be the case.

Most of us reckon we can largely say if someone is under 25 using the Challenge 25 technique but when put to the test, the truth comes out: retailers do let some under 18s buy alcohol. Not always and not everyone, but some people are incorrectly estimated to be older than they really are. Let’s be honest, this is not ideal.

Now, to be fair, not all humans are created equal. Some people are fairly accurate at guessing another person’s age. A few are even impressively accurate. We found one recently. His name is Elliot, one of the human estimators who took part in a recent experiment with The Times. His main career is currently in professional sport, but he occasionally undertakes work as a Super Recogniser.

But even Elliot – one of the best human estimators – isn’t as accurate as Yoti’s facial age estimation technology.

So here’s the question: do we just need to clone Elliot? Stick him in every supermarket, corner shop, bar, and petrol station across the UK in order for human estimation to be a more accurate process than it currently is?

Let’s do some silly but very real maths:

  • There are around 224,100 licensed premises in the UK.
  • Let’s say we pay Elliot £25k a year (a bargain for an elite-level age estimator).
  • If Elliot is full time age estimating, we now have a £5.6 billion-a-year wage bill… just to check people’s ages.

And that’s assuming Elliot never gets tired. Never calls in sick. Never gets flustered when a queue of angry customers forms because he had to triple-check one customer whose age he was less certain of.

Meanwhile, Yoti, headquartered in the UK, already has technology that can do this job. That does it to a very high degree of accuracy. It’s fast, consistent, tested, and doesn’t get flustered or tired.

Retailers want to use it. They’ve already tested it. Staff love it – mostly because it means they don’t have to awkwardly guess whether someone is 17 or just has a baby face. And customers like it too.

But here we are, in 2025, still relying on busy, stressed-out staff and the lower accuracy of human estimation to prevent underage sales.

The technology has been sitting there on the shelf, ready to go for a few years. Surely it’s time to stop pretending that retail workers trying to guess the age of customers is the gold standard in age assurance, especially since the government says it wants to see tech make people’s lives easier.

Let’s allow businesses to give people the choice to use facial age estimation, the tool that works really well for this job.

Keep reading

Woman presenting a 2d image trying to perform a presentation attack
Articles June 19, 2025

Why early detection is critical in stopping deepfake attacks

Digital identity and age verification are becoming integral parts of customer onboarding and access management, allowing customers to get up and running on your platform fast. However as customer verification tools become more advanced, so too are fraudsters seeking to spoof systems by impersonating someone, appearing older than they really are or passing as a real person when they’re not. Deepfake attacks, which can mimic a person’s face, voice or mannerisms, pose a serious threat to any business using biometric customer verification. In this blog, we explore why detecting deepfakes early is essential for maintaining trust, security and regulatory

#deepfake attacks#business
Elly Heath Elly Heath
6 min read
Woman using facial age estimation technology at a self-checkout
Articles June 16, 2025

Why facial age estimation, the most accurate age checking tool, shouldn’t be left on the sidelines

Many of us have been there: standing at a self-checkout, scanning our shopping, only to hit a roadblock when the till flags an age-restricted item like a bottle of wine or a pack of beer. With age verification accounting for between 40 – 50% of interventions at self-checkouts, it significantly disrupts and slows down the checkout experience. We wait for a retail worker to approve the sale. The retail worker does a visual estimation of our age – they look at our face and guess whether we’re old enough to buy the item. Most retailers follow the Challenge 25

#age assurance#business
Rachael Trotman Rachael Trotman
6 min read
Woman at desk using multiple screens
Articles June 9, 2025

Why testing data is as important as training data for machine learning models

When developing machine learning systems for facial age estimation, the conversation often centres on the training data: how much you have, how diverse it is, how inclusive it is, and how well it represents your end users.  Not to mention, where the data comes from.  Intuitively, that focus makes sense. More data presumably leads to better models. But test data is just as important, and in some ways, even more critical for ensuring models perform effectively. Training data: more isn’t always better Common sense would suggest that for a machine learning model “the more data, the better.” And that’s

#age assurance#business
Matt Prendergast Matt Prendergast
4 min read