Breaking AI & Tech News Analyzed

California is taking a significant step in regulating the artificial intelligence (AI) industry by introducing new standards for companies seeking to do business with the state. This move directly contradicts former President Donald Trump's stance on keeping the industry as deregulated as possible. Governor Gavin Newsom signed an executive order on March 30, giving the state four months to develop AI policies that prioritize public safety. Companies hoping to secure contracts with California will be required to demonstrate policies that prevent AI from distributing child sexual abuse material and violent pornography. They must also show how their models avoid incorporating “harmful bias” and detail policies aimed at avoiding “unlawful discrimination, detention, and surveillance”. The order also directs the state to come up with best practices for watermarking AI-generated or -manipulated images and videos. Newsom emphasized California's commitment to innovation while ensuring that companies protect people's rights and do not exploit or put them in harm's way. California's actions are part of a broader trend of state-level attempts to regulate an AI industry that has raised public safety concerns and worries about the potential for job displacement due to automation. According to the New York Times, states have passed more than 100 laws to shield children from chatbots and to block AI companies from using copyright-protected material. The White House issued a national policy framework for AI in December that discouraged states from passing such regulations, with Trump's executive order calling for minimal regulation to allow U.S. AI companies to innovate freely. In response, the Justice Department established an “AI Litigation Task Force” to challenge state AI regulations.

A New Mexico jury has ordered Meta to pay $375m in civil penalties after finding the company liable for misleading consumers about the safety of its platforms and enabling harm, including child sexual exploitation. This verdict marks the first bench trial to find Meta liable for acts committed on its platform. The lawsuit, brought by New Mexico Attorney General Raúl Torrez, claimed that Meta executives knew their products harmed children but disregarded warnings from their own employees and lied to the public about the risks. The jury found Meta liable for violating New Mexico's consumer protection laws, specifically the Unfair Practices Act. The penalty of $375m is the maximum allowed under the law of $5,000 per violation. Meta has announced its intention to appeal the ruling, accusing Torrez of making 'sensationalist, irrelevant arguments.' Internal Meta documents and testimony revealed that company employees and external child safety experts repeatedly warned about risks on Meta's platforms. Evidence presented included details of a sting investigation, 'Operation MetaPhile,' which led to the arrest of three men charged with sexually preying on children through Meta's platforms. The New Mexico court also heard about deficiencies in Meta's reporting of crimes on its platforms, including the exchange of child sexual abuse material. Meta generated high volumes of 'junk' reports by overly relying on AI to moderate its platforms, making it difficult for law enforcement to investigate crimes. In the next phase of the legal proceedings, the attorney general's office will seek additional financial penalties and court-mandated changes to Meta's platforms to offer stronger protections for children. The state is seeking design feature changes, including 'enacting effective age verification, removing predators from the platform, and protecting minors from encrypted communications.' This verdict is seen as a 'historic victory' for children and families who have been affected by Meta's actions. The case may also open the floodgates to further litigation and regulation of social media companies.

The amount of AI-generated child sexual abuse material found online surged by 14% in 2025, according to the Internet Watch Foundation (IWF). The watchdog identified 8,029 AI-made images and videos of realistic child sexual abuse material (CSAM) last year.The IWF reported a 260-fold increase in videos, with 65% of the 3,443 videos classified as category A, the most severe type under UK law. This is compared to 43% of non-AI videos. Kerry Smith, IWF chief executive, emphasized that advances in technology should not come at the expense of a child's safety and wellbeing.The IWF also found that offenders are discussing the potential of AI systems to create more realistic and violent content, including adding audio to videos or manipulating imagery of real children. The UK government has announced plans to test AI tools to prevent CSAM and introduced legislation to ban AI models designed to generate child sexual abuse material.Eight out of 10 UK adults want the government to introduce legislation ensuring AI systems are developed with safety as a priority. The IWF operates a hotline and has a global remit to monitor child sexual abuse content.