The New York Supreme Court allowed a lawsuit against TikTok to proceed after a bipartisan coalition of U.S. state attorneys general claimed the platform's algorithms negatively impact users' mental health.
The U.K. AI Security Institute published research evaluating frontier AI model safeguards and how to make those evaluations actionable for developers. The analysis explores the metrics that should be used to measure safeguard efficacy and the thresholds the metrics should be aligned with for a given risk level.Full story
The International Organization for Standardization and International Electrotechnical Commission released their joint ISO/IEC 42005 standard for conducting AI impact assessments. The standard complements existing AI standards with a focus on transparency, accountability and trust. It seeks to help organizations align development with values, such as fairness, safety and human-centered design for AI models.Full story
The White House is expanding the work of data analysis firm Palantir across the U.S. government after President Donald Trump signed an executive order in March calling for federal agencies to share data across their operations, The New York Times reports. Palantir's Foundry product has been integrated with "at least" four federal agencies, including the Department of Homeland Security and Department of Health and Human Services.
The U.S. District Court for the Southern District of New York ruled employees at the Department of Governmental Efficiency may access Department of Treasury data if properly vetted and trained, The Washington Post reports. The government was sued numerous times over DOGE's attempts to access U.S. sensitive information without going through typical department channels.Full story
Disagreements over what counts as a solely automated decision and how to reconcile the right to information while protecting proprietary information were major themes in the public feedback on Brazil's data protection authority's, Autoridade Nacional de Proteção de Dados, discussion on AI and automated decisions, writes Tiago Neves Furtado, CIPP/E, CIPM, CDPO/BR, FIP, Opice Blum Advogados' lead on data protection, AI and the incident response team.
Vietnam has undergone significant reforms to its data protection framework, with several regulations passing in the last three years and at least five more expected in 2025, Baker McKenzie Vietnam's Alex Do writes. He maps out the regulatory infrastructure of Vietnam's data protection regime while previewing what the forthcoming regulations may contain.Full story
In an op-ed for Tech Policy Press, Georgetown University adjunct professor Mark MacCarthy said data collected by AI technology may allow users to have personalized experiences when using the services, though the technology could have privacy implications.
The Vermont Legislature granted final passage to Senate Bill 69, the Age-Appropriate Design Code Act. The bill mandates default privacy and safety settings for children under 18 and requires online content platforms to restrict certain content and turn off addictive features. If enacted by the governor, the bill will take force 1 Jan. 2027.Full story
As part of its 25th anniversary, the IAPP celebrates 25 trailblazing innovators and 25 defining moments in the last quarter century that helped establish the privacy and digital governance profession. This week, we highlight the pioneering work of Nuala O'Connor, CIPP/G, and her influence on the broader public and private-sector privacy communities.Full story