Archive · 2 Jun 2025
Meta to Replace Human Reviewers With AI for Risk Assessments
Meta, the company behind Facebook and Instagram, plans to use artificial intelligence (AI) to check if new features are safe.
Instead of people, AI will handle up to 90% of the reviews, making decisions about things like privacy and youth safety.
The company hopes this will help them release new products faster.
Some people worry that reducing human reviewers could lead to problems.
Meta says humans will still review complicated issues and AI will handle low-risk decisions.
One employee said that this is a bad move.
Meta is shifting to AI for risk assessments of its products.
AI will handle up to 90% of internal risk assessments.
The change affects areas such as algorithms and youth safety.
The goal is to increase the speed of feature and product releases.
Concerns exist about the reduction of human oversight.
- Who
- Meta
- What
- Meta is replacing human reviewers with AI for risk assessment of its products and features.
- When
- The transition reportedly started in April and continued through May 2025.
- Where
- Meta's global operations, with potential exceptions.
- Why
- To speed up product development and streamline the review process.
Advocates of AI Automation
Critics of AI Automation
Speed vs. Accuracy
Advocates of AI Automation
AI feedback will be near-instant, accelerating product releases.
Critics of AI Automation
The shift could lead to reduced scrutiny and potential risks.
Human Oversight
Advocates of AI Automation
Meta believes AI can handle low-risk decisions effectively.
Critics of AI Automation
Current and former Meta employees are concerned about removing human perspectives.
- Company
- Meta (Facebook, Instagram, WhatsApp, Threads)
- Change
- AI-powered automation for risk assessments
- Automation Level
- Up to 90%
- Areas Affected
- Algorithms, safety features, content sharing, youth safety, AI ethics
- Company Goal
- Speed up product releases
