Under the hood of facial age estimation

profile picture Matt Prendergast 4 min read
An image showing a woman using her mobile phone. An illustration shows that the owner of the account matches the person who attempting to log into it.

Since the UK Online Safety Act came into force on Friday 25th July, there has been a lot of discussion about how effective age checks are. 

Facial age estimation has been approved by Ofcom as a high assurance method for online age checks. It’s a quick, private and effective way to confirm if someone is above or below an age threshold. 

Each facial age estimation is quick and simple – taking around one second. This has sparked conversation when talking to businesses, press, regulators and governments – one of the most striking comments we’ve heard when showing them the technology is: “is that it?”. 

Well, yes – but in that second, there’s an awful amount happening behind the scenes, with our anti-spoofing technology playing a critical role.

What happens to complete a facial age estimation check

There are broadly four important steps to a facial age estimation check:

  • Image capture
  • Age estimation
  • Anti-spoofing checks
  • Image deletion

Image capture

The first step to a successful age assurance check is a good quality image. Yoti has an automated face capture module that prompts the user to frame their face for an optimal result. There is then no further action required of the user, like further movements, or the requirement to repeat words. Face capture automatically captures the best image for a successful age check. This improves accessibility. 

Age estimation 

We then break down that image into pixels – which is essentially a series of numbers. We can then analyse that with a mathematical model using our machine learning algorithm, which is able to produce a reliable age result from that data. We return only the result required, for example, ‘over 18’. We do not know who the person is, and neither does the platform.

Anti-spoofing checks

An equally important question is, how secure is the process? There’s little point in estimating the age of a face accurately if it’s not a real face in front of the camera. This is why it is important to use a combination of technologies to offer a high level of assurance. 

An attempt to present an image, mask or video is often called a presentation attack. To overcome this we use our NIST 2 certified liveness technology. This ensures that the person undertaking the check is a real person and not someone wearing a mask, or presenting a picture or video of another (older or younger) person to the camera. 

Image deletion

The image of the user is then deleted and no data is stored. The only thing we pass over to the platform is a ‘Yes/No’ answer as to whether the person meets the required age threshold or not. It is impossible for the platform, and for Yoti, to know the identity of the person from facial age estimation

Why use facial age estimation

Facial age estimation is a practical and effective way of ensuring children can enjoy an online experience without being able to access inappropriate content. We have successfully performed over 850 million age checks to date, the equivalent to a tenth of the global population.  

Keep reading

How accurate can facial age estimation get?

Facial age estimation using machine learning has advanced significantly in recent years. But, a common and fair question still arises: How accurate can it really be? Can a system look at your face and accurately guess your age, especially when humans often get it wrong? The short answer is that it’s very accurate – but not perfect. We explain why.   The myth of 100% accuracy It’s important to set realistic expectations. No facial age estimation model can achieve 100% accuracy across all ages.  Human aging is highly individual and shaped by many external factors, especially as we get

6 min read
An image of a woman trying to buy a bottle of alcohol at a supermarket self-checkout terminal.

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

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.

3 min read
Woman using facial age estimation technology at a self-checkout

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

6 min read