Inc.
Age checks may protect teens from harmful AI and social platforms, but they could create a new privacy nightmare.
The Next Internet Crisis Might Start With Age Verification
Author: Heather Wilde Renze
Online age verification has been an issue for more than a decade—and one that has not yet found a perfect solution. As social media and AI come under increasing scrutiny for their impact on teens and children, age verification has become even more relevant.
Generative AI, in particular, is creating new challenges. Teens often use AI as a companion without parental awareness. With as many as 19 percent of teen users being encouraged to participate in dangerous behaviors by a chatbot, it’s understandable why many believe age verification is necessary to protect teens from potential harm online.
However, the most cited solutions for age verification present their own challenges, which could fuel the next digital trust crisis.
Why are age-related concerns rising?
Age verification concerns have grown as generative AI has proliferated and shown significant potential for harm among minors. A World Economic Forum report cites an explosion of threats, such as AI-generated imagery of child sexual abuse and deepfake nudes, which contribute to harassment and extortion.
At the same time, over-trust of AI and emotional dependency are creating issues among teens and adults alike. Minors are more likely to disclose sensitive information. They might even follow harmful advice delivered by a chatbot.
Most sobering are the instances of teens who died by suicide or harmed themselves after talking to a chatbot. Google and Character AI have settled multiple lawsuits related to such incidents.
For many, these drastic outcomes require the equally drastic solution of age verification. This ensures that minors can’t gain access to potentially harmful platforms. But AI age verification is rife with its own challenges.
Current solutions are far from ideal
Right now, the growing trend is to use driver’s licenses or other government IDs to verify user ages. This is largely in response to new laws designed to limit kids’ access to social media and generative AI.
While the regulations are well-intended, the resulting verification solutions are far from ideal. Some areas require age verification for a broader range of websites. This can sometimes limit access to news and political topics.
As Proton founder and CEO Andy Yen writes, current AI age verification practices can create new hazards. This is heavily due to the large quantity of sensitive data that is collected to verify individuals’ ages and identities.
Apps and even governments are surprisingly vulnerable to hackers, and they only become more appealing targets as they collect more identity verification data.
The current models that require users to send in fingerprints or photos of government IDs lead to new risks of data exposure. Yen argues that solutions like facial scans that are deleted after they are processed are a better solution.
Quite simply, no one wants to hand over their driver’s license to access TikTok. Yet that is what many of the current age verification standards propose. And it’s not just teens being asked to hand over information—adults also need to verify their age.
The rise of deepfakes is another relevant concern. If data uploaded for age verification is compromised, bad actors have even easier access to a wide swath of information. This access allows them to steal a person’s likeness to create accounts, post harmful content using that person’s identity, make fraudulent purchases, and more.
As more websites that collect sensitive information are hacked and data is exposed, it can ultimately create a new internet trust crisis. Such measures could do far more than just drive people away from social media or generative AI tools. It could deter many people from accessing information resources that they need.
Finding appropriate verification solutions
Finding the right balance in AI age verification is rife with obstacles, particularly with the data-heavy approach that is currently favored. However, it’s clear that there are other solutions that can pose less of a data trust concern.
In line with Yen’s suggestion of quickly deleted facial scans, New America recommends comprehensive data privacy legislation. This will minimize the amount of data companies can collect online. It will also place restrictions on what they can do with that data. Paired with greater platform transparency and standardized third-party verification solutions, platforms can find a better balance in protecting youth. They can also minimize the potential negative outcomes of age verification.
Ultimately, as more groups push for age verification, it becomes increasingly important that the verification methods themselves are also subjected to greater scrutiny. Otherwise, giving up privacy and sensitive data has the potential to create new harms and further erode online trust when that information is stolen or otherwise misused.
Credits: TCA, LLC.