California's Landmark AI Transparency Law Took Effect August 1

Nation's first comprehensive AI transparency framework ushers in a quiet revolution of digital transparency and consumer protection

SACRAMENTO, CA — On August 1, California began implementing the California Artificial Intelligence Transparency Act (CAITA), the nation's first comprehensive AI transparency framework, ushering in what Senator Josh Becker (D–Menlo Park) calls "the beginning of a quiet revolution" for the internet. 

You can view the announcement here. 

While websites, social media platforms, and online content will look much the same, a fundamental change will begin beneath the surface. AI-generated images, videos, and audio will increasingly carry tamper-evident provenance information that helps consumers understand where digital content came from and whether it has been created or manipulated using artificial intelligence.

This new law comes as AI-generated content poses growing challenges for our democracy and the 2026 election. Highly realistic deepfakes, manipulated images, and synthetic audio can be used to impersonate candidates, public officials, or election workers, making it harder for voters to distinguish authentic information from fabricated content. While transparency alone cannot prevent election misinformation or fraud, it gives voters and journalists another tool to verify the origin of digital content and make more informed decisions about what they see and share.

"As artificial intelligence becomes more powerful and more widespread, voters and consumers deserve better tools to understand what they're seeing online," said Senator Becker. "Beginning August 1, California starts building a more transparent internet. Most people won't notice dramatic changes overnight, but the information behind digital content will begin changing in profound ways to protect consumers."

"The CAITA landmark legislation played an instrumental role in helping us define and negotiate the rules for marking and detection of AI-generated and manipulated content, published as part of the European Code of Practice on Transparency of AI-Generated Content. The European marking and detection rules are aligned with CAITA's provisions and will similarly go into effect from August 2nd 2026. We are looking forward to improved content transparency in both jurisdictions," said the chairs of the European working group one on the Code Prof. Kalina Bontcheva, Prof. Dino Pedreschi, and Prof. Christian Riess.

“The California AI Transparency Act is a landmark victory for digital safety and consumer trust," said Jai Jaisimha, co-founder of the Transparency Coalition. "By establishing clear rules for provenance and disclosure, California is proving that we can foster groundbreaking technological innovation while fiercely protecting the public from deception. We have been proud to support Sen. Becker and others in the CA legislature in crafting and passing this landmark legislation and look forward to its full implementation."

As CAITA is phased in over the coming months, AI-generated content will increasingly carry provenance information that remains attached as it is shared online. Consumers will be able to use publicly available tools to inspect that information and determine where content originated and, in many cases, who created it. While the technology will continue to mature, these disclosures are expected to become increasingly detailed and resilient over time.

The next major phase of implementation begins on January 1, when large online platforms will be required to make provenance information more accessible by allowing users to inspect AI-generated content directly within their platforms. Beginning in 2028, qualifying capture devices will also begin embedding provenance information into the content they create, providing additional information about when, where, and how digital content originated.

"Together, these protections will fundamentally change how we interact with information online," Becker said. "The goal isn't to tell people what to believe. It's to give them better information so they can decide for themselves who and what to trust."

The law comes as AI-generated content becomes increasingly sophisticated. Artificial intelligence is making it easier for scammers to impersonate loved ones, clone voices, fabricate convincing images and videos, and launch highly sophisticated phishing attacks. Older adults are especially vulnerable to these scams, which cost Americans more than $1 billion annually.

CAITA represents a major step toward making deception more difficult by providing consumers with greater transparency about AI-generated content.

California's implementation is also drawing international attention. Senator Becker has worked extensively with technology companies, academic experts, standards organizations, consumer advocates, and international policymakers—including officials helping develop the European Union's AI Code of Practice and French Ambassador for Digital Affairs Clara Chappaz—to help advance interoperable global AI transparency standards.

"In the absence of federal leadership, California is demonstrating that innovation and consumer protection can move forward together," Becker said. "The world is watching how we implement AI transparency, and we're committed to ensuring our laws continue evolving alongside this rapidly changing technology."

To help keep California's framework current, Senator Becker has authored SB 1000, legislation that strengthens interoperability with emerging international standards, addresses implementation challenges, and ensures the state's landmark AI transparency law continues to evolve as AI technology advances.