Meta plans to automate many of its product risk assessments | TechCrunch
May 31, 2025 - techcrunch.com
Meta is reportedly planning to implement an AI-powered system to evaluate the potential harms and privacy risks of up to 90% of updates made to its apps, such as Instagram and WhatsApp. This move comes in response to a 2012 agreement with the Federal Trade Commission that requires Meta to conduct privacy reviews of its products. Traditionally, these reviews have been conducted by human evaluators, but the new system will involve product teams filling out a questionnaire and receiving an "instant decision" from the AI regarding identified risks and necessary requirements for updates.
While this AI-centric approach aims to expedite product updates, it has raised concerns about increased risks, as potential negative impacts of product changes might not be identified before they cause issues. Meta has confirmed changes to its review system, stating that only "low-risk decisions" will be automated, while human expertise will continue to address "novel and complex issues."
Key takeaways:
Meta plans to use an AI-powered system to evaluate up to 90% of updates for apps like Instagram and WhatsApp.
The AI system will provide instant decisions on potential risks and requirements for updates, based on a questionnaire filled out by product teams.
This approach aims to speed up product updates but raises concerns about higher risks due to less human oversight.
Meta claims that only low-risk decisions will be automated, with human experts handling more complex issues.
