> **Meta to Automate 90% of AI Risk Assessments**
Meta is reportedly planning to automate up to 90% of AI risk assessments for updates across its platforms, including Instagram and WhatsApp, according to internal documents reviewed by NPR. The move would shift responsibility from human reviewers to AI-driven evaluations, significantly speeding up the development process.
### How the New AI Risk Assessment System Works
Under the new system, product teams will fill out a questionnaire about their work, and an AI-powered tool will generate an instant risk assessment. The AI will flag potential privacy concerns and regulatory compliance issues before updates are deployed.
This shift is part of Meta’s broader strategy to accelerate product updates while maintaining compliance with regulatory requirements. However, critics warn that relying on AI for risk assessments could lead to unintended consequences, as automated systems may miss nuanced ethical or privacy concerns that human reviewers would catch.
### Regulatory and Ethical Concerns
The change comes as Meta continues to face scrutiny over data privacy and AI governance. A 2012 agreement with the Federal Trade Commission (FTC) requires Meta to conduct privacy reviews for all product updates. While the company claims its AI system will enhance efficiency, some former executives argue that automating risk assessments could increase the likelihood of harmful updates slipping through.
### Meta’s Response
In a statement, Meta emphasized its $8 billion investment in privacy programs and its commitment to balancing innovation with regulatory compliance. The company stated:
> “We leverage technology to improve consistency in low-risk decisions while relying on human expertise for complex or novel challenges.”
The shift toward AI-driven risk assessments reflects a growing trend in the tech industry, where companies increasingly rely on machine learning to streamline operations. However, the long-term impact of this approach on user privacy and platform safety remains to be seen.
(This article has been updated with additional details from Meta’s official statement.)