With more regions considering increased teenage social media bans, and public support for broader bans gaining traction around the world, European policymakers are meeting in Brussels this week to discuss next steps, and what the best way forward might be for age restrictions online.
And Google will be in attendance, with the search giant today outlining its view on age checking, and the challenges of verifying user ages in a safe and effective way.
Which is says is generally pitched “as a false choice between weak age gates and invasive ID scans.”
Google says that people largely understand the need for stronger protections, “but they also don’t want their IDs exposed in a data breach.”
In Google’s view, however, the invasiveness of such approaches should be relative to the risk.
As per Google:
“Our research supports a ‘risk-based’ approach where assurance matches risk. That means less intrusive assurance methods in most areas – news, education, or travel – and stronger checks for things like adult content or alcohol sales. The rigour matches the risk.”
So, yes, you should have to provide ID, but not for everything, just for the more sensitive elements. Which I doubt that anybody will disagree with, but the broader debate here is less about the actual process, and more the method with which age checking can be implemented at scale.
On this front, various solutions have been proposed, including video selfie verification, machine-learning models, etc. There’s no universally agreed best approach, though again, limiting data exposure is a key consideration, because as noted by Google, invasive ID scanning leads to potential risk.
Though Google has also added this somewhat bizarre justification to alleviate its own responsibility to implement age checking at the app store level:
“While some would like a more uniform system, or a universal arbiter of age, we believe that liability and responsibility rest with every service owner – the developer, the publisher, the app creator – because they know what they are offering. You don’t expect the credit card company to check if you are old enough to buy alcohol; the store should do that.”
Okay, sure, you don’t expect the credit card company to check for ID at the point of sale. But then again, it would make things much easier if they could.
For example, if there was a universal age-checking system that would stop an underage user from buying alcohol with a credit card, based on a centralised system that didn’t require manual age checking in store, that would make it much easier for retailers, while also limiting underage access at scale.
That would be a better solution, right?
Of course, we can’t do that with alcohol, but we can with apps, where Google and Apple control the access point, and can verify user ages across all apps at the download stage. That would mean more universal enforcement, and less room for data leaks, due to multiple age-checking approaches in every app.
But it would also put more onus on both Apple and Google to check user ages, and leave them open to potential liabilities as a result of mistakes. As such, it’s no surprise that Google’s like “it would be dumb for us to check ages,” but as Meta has noted many times, this would alleviate many of the concerns about age checking, by limiting data access:
“By verifying a teen’s age on the app store, individual apps would not be required to collect potentially sensitive identifying information. Apps would only need the age from the app store to ensure teens are placed in the right experiences for their age group. Parents and teens won’t need to provide the hundreds of apps their teens use with sensitive information like government IDs.”
Which is the exact issue that Google has highlighted, that concerns around potential privacy concerns limit some age-checking options.
So limiting the amount of times such data needs to be entered would be a better way, right?
Evidently, and logically from a business perspective, Google doesn’t agree, though as we’re seeing in Australia, which is close to launching its own teen social media restriction laws, other, platform-specific methods of checking user ages are ultimately going to prove less effective.
Meta has now begun notifying Australian teens that it will soon be implementing more stringent age-checking processes, as required by the new law, which comes into effect on December 10th.
Though as reported by Bloomberg:
“But [Meta has] cautioned that there remained a ‘significant’ margin of error when determining whether a user is 16 years or older, and it expects to misidentify an unspecified number of underage users, as well as those who are legally allowed to hold accounts.”
A key flaw in the Australian government’s approach here is that it hasn’t dictated a preferred method of age checking in order to abide by the new law, it’s simply noted, through its own trials of a range of age checking options, that there are adequate age verification tools available that will enable social platforms to adhere to the new requirements.
The law itself states that all social media platforms will have to “take reasonable steps” to restrict teens under the age of 16 from accessing their apps.
But “reasonable” leaves a lot of legal wiggle room, and without a prescribed, allocated tool that every platform has to implement to adhere to these new requirements, it’s hard to see how local authorities will be able to enact penalties based on this.
In the end, then, I suspect that the impacts of the change will be minimal, with teens working out how to side-step the various measures, and the platforms arguing that they are indeed undertaking “reasonable steps” to stop them, whether they work or not.
We’ll find out soon, with Meta now informing Australian teens under 16 that they’ll have 14 days to access their accounts, before they’re cut off on December 10th, while EU leaders will soon be voting on new approaches to cover their region.
And you can bet that all regulators will be watching on as Australia’s law comes into effect.
