AI News Roundup — March 19, 2026
The White House released a national AI policy framework to pre-empt state regulations, Nvidia GTC 2026 wrapped up its record-breaking four-day run, and the US passed the AI Accountability Act requiring bias audits for consequential AI decisions.
White House Releases National AI Framework to Pre-Empt State Rules
The Trump administration released a national AI policy framework aimed at ensuring protections for children, communities, and small businesses. The plan pushes for a single federal legislative framework applied uniformly across the country, rather than allowing individual states to form their own AI regulations — a move that has drawn both support from industry and criticism from consumer advocates.
Nvidia GTC 2026 Wraps Up After Record-Breaking Four-Day Run
GTC 2026 closed on March 19 with a campfire gathering of robots and Toy Jensen recapping the show's highlights. The final day featured a packed session with Nvidia chief scientist Bill Dally and Google DeepMind's Jeff Dean discussing the future of large-scale machine learning. The conference drew 30,000 attendees from 190 countries — 20% more than last year — and showcased 110 robots across its venues.
MiniMax M2.7 Model Launches with Autonomous Agent Capabilities
MiniMax released M2.7, a new model now publicly available via the MiniMax Agent and API Platform. The model supports complex workflows in software engineering, office productivity, and research environments, with standout capabilities including autonomous debugging and research agent harnesses for multi-step task execution.
Anthropic Study of 80,000 People Reveals Public Hopes and Fears About AI
Anthropic conducted a large-scale study with 80,508 global participants to understand public sentiment toward AI. Key desires include professional excellence and life improvements, while main concerns center on AI unreliability and job displacement. The study found that people experience AI as both a productivity tool and a potential dependency, offering nuanced insights for AI developers and policymakers.
Google Brings Gemini to Docs, Sheets, Slides, and Drive
Google announced new Gemini features across its Workspace suite, enabling users to write documents, create spreadsheets, design presentations, and find information across files and emails using natural language. The AI-powered features are available for Google AI Ultra subscribers and represent a significant step in embedding AI throughout everyday productivity tools.
US Passes AI Accountability Act Requiring Bias Audits
The United States passed the AI Accountability Act in March 2026, requiring companies deploying AI in consequential decisions — including hiring, lending, healthcare, and criminal justice — to conduct and publish regular bias audits. The landmark legislation ends years of voluntary self-regulation and establishes the first comprehensive federal oversight framework for AI decision-making.
McKinsey Report Finds 12% of Job Tasks Automated but Net Employment Flat
McKinsey's March 2026 report found that 12% of current job tasks have been automated by AI over the past two years, but 8% of new job categories created were directly AI-related. Net employment in AI-exposed sectors is approximately flat, offering a more nuanced picture than the dramatic disruption forecasts from other analysts and suggesting AI is reshaping work rather than simply eliminating it.