Archive — US — techcrunch
Meta plans to automate many of its product risk assessments
Meta plans to automate up to 90% of product risk assessments for apps like Instagram and WhatsApp using AI, according to internal documents. This shift, driven by a need for faster updates, raises concerns about potential risks and negative consequences, despite Meta's assertion that only low-risk decisions will be automated while human oversight remains for complex issues.
Key points
- Meta is implementing an AI system to assess risks in app updates.
- This change affects up to 90% of updates for apps like Instagram and WhatsApp.
- Currently, evaluations are conducted by humans as per a 2012 FTC agreement.
- The new system involves teams filling out questionnaires for risk assessment.
- AI will provide instant decisions and outline necessary compliance requirements.
- While aimed at speed, critics warn it may lead to increased risks from unexamined updates.
- Meta claims that only low-risk assessments will be automated, with human reviews still for complex issues.
Why it matters
- This development aligns with the growing trend of AI automation in various industries as of 2025.
- Similar concerns about AI's role in accountability and oversight have been raised in other tech sectors.
- Increased automation raises debates about risk management and consumer safety in tech innovations.
In summary, Meta's shift towards AI for product risk assessments illustrates a dual focus on efficiency and safety, yet raises important questions about the adequacy of oversight in the tech landscape.