Zu den neuen ChatGPT-Sicherheitsmaßnahmen für Jugendliche gehören Altersvorhersage und Altersüberprüfung
https://www.aol.com/news/chatgpt-teen-safety-measures-age-152732928.html
Zu den neuen ChatGPT-Sicherheitsmaßnahmen für Jugendliche gehören Altersvorhersage und Altersüberprüfung
https://www.aol.com/news/chatgpt-teen-safety-measures-age-152732928.html
5 Kommentare
lol that’s going to be interesting…
we are about to find out what adult is intellectually a teen and what teen is intellectually an adult based on algorithms….
Next step… selling that data. it won’t be long before AI is too restrictive to be useful. Then.. the next boom.. „less filtering on A & B, more freedom on C at an extra cost“.
We are „THIS“ close to an LLM black-market
> „If there is doubt, we’ll play it safe and default to the under-18 experience,“ Altman wrote. „In some cases or countries we may also ask for an ID; we know this is a privacy compromise for adults but believe it is a worthy tradeoff.“
Asking for ID, selfies, and videos is a $$$$$$ compromise that enriches age verification providers who profit from violating user privacy.
If they’re using AI to verify ages then it probably isn’t going to end well. When I was 16 I could just go without shaving and easily pass as 30.
The „age prediction“ aspect is arguably more dystopian than the ID check. To accurately predict age based on usage, they essentially have to profile your syntax, vocabulary, and the topics you discuss against a „teen“ baseline.
It creates a bizarre paradox: to protect minors, you have to surveil everyone’s private thoughts to ensure they sound „adult“ enough. It also begs the question of false positives — how many adults with casual typing styles or niche interests are going to get flagged and forced to hand over an ID? This feels less like safety and more like the normalization of behavioral biometric profiling.
Age checks help keep teens safer. That is the clear win. The risk is feature creep and data misuse. Companies push for safety, then slowly turn it into tracking or paid access. People worry about that pattern. If the tools stay transparent and limited, they help. If not, they turn into another system that watches users more than it protects them.