Content Moderation
Cloudflare introduces Guardrails in AI Gateway to help developers deploy AI applications safely by monitoring and controlling content through Llama Guard integration. The feature addresses challenges of inconsistent safety measures across AI models and provides comprehensive visibility into user interactions while helping meet regulatory requirements. Guardrails offers granular control over content moderation, allowing developers to flag or block inappropriate content based on predefined hazard categories.
The UK's Online Safety Act, effective March 16, 2025, requires strict compliance from websites with UK visitors, regardless of location or ownership. Lobsters, a non-commercial hobbyist forum, faces potential geoblocking of UK users due to inability to meet complex regulatory requirements and potential $22M penalties.
Brazilian Supreme Court Justice Moraes ordered video platform Rumble to be banned in Brazil for non-compliance with court orders, including refusing to take down a controversial journalist's channel. Rumble, which powers Trump's Truth Social platform, has filed a lawsuit against Moraes, claiming unprecedented censorship and violation of US constitutional rights.
A user shares their negative experience with LinkedIn, detailing multiple account restrictions, shadow-banning incidents, and verification challenges within their first three months of usage. The platform's algorithmic bias and poor treatment of new users, regardless of premium status, highlights significant user experience issues.
Multiple social media platforms are implementing measures to counter disinformation ahead of Germany's federal election, following concerns about Russian interference and manipulated content. TikTok, Meta, Snap, and Google have reported removing fraudulent campaigns, launching election information centers, and enhancing content moderation systems.