The artificial intelligence landscape is moving at a breakneck pace, and lawmakers are struggling to keep up. What started as experimental chatbots and image generators has rapidly evolved into foundational technology embedded in healthcare, finance, education, and everyday consumer apps. In this shifting environment, one of the industry’s most prominent voices is making a clear call to action: state governments need to move faster on AI regulation. Anthropic, the AI research company known for its focus on safety and transparency, is actively pushing state legislatures to update their regulatory frameworks, arguing that even recently passed laws may already be falling behind the technology they were designed to govern.
The Speed of Innovation Versus the Pace of Law
Legislation typically moves in slow, deliberate cycles. Drafting, debating, amending, and passing laws can take months or even years. Artificial intelligence, on the other hand, iterates in weeks. Models are updated, capabilities expand, and new use cases emerge almost overnight. This mismatch creates a significant challenge for policymakers who want to protect consumers without stifling innovation. Anthropic’s leadership has pointed out that the regulatory frameworks being debated today are often based on yesterday’s technology. By the time a bill clears committee, the industry landscape may have shifted entirely, leaving lawmakers with rules that no longer align with current realities.
Why State-Level Regulation Is Taking Center Stage
With federal AI legislation stalled in Washington, individual states have stepped into the void. States like California, New York, and Colorado have become testing grounds for how to balance consumer protection with technological progress. These jurisdictions have the flexibility to experiment with different approaches, creating a patchwork of regulations that could eventually inform national standards. Anthropic’s head of U.S. state and local policy has emphasized that state governments are uniquely positioned to respond quickly to emerging risks. However, that responsiveness only works if the underlying laws are built to adapt. Static rules that lock in specific technical requirements risk becoming obsolete the moment a new model architecture or training method gains traction.
The California and New York Precedent
Last year, Anthropic publicly endorsed landmark AI transparency laws in both California and New York. These bills aimed to bring greater accountability to how AI systems are developed, deployed, and monitored. They required companies to disclose training data sources, implement safety testing, and maintain clear documentation of model behavior. At the time, these measures represented a significant step forward. But as the company has continued to work closely with state regulators, it has become clear that even these forward-thinking frameworks may need rapid updates. The pace of model development, the emergence of agentic AI systems, and the growing complexity of AI integration into existing software ecosystems have introduced new variables that earlier drafts simply didn’t account for.
The Case for Adaptive, Future-Proof Policies
Rather than prescribing rigid technical mandates, Anthropic is advocating for adaptive regulatory frameworks. This approach focuses on outcomes rather than specific methodologies. Instead of dictating exactly how a company must test for bias or prevent data leakage, regulators can set clear safety and transparency benchmarks while allowing companies the flexibility to meet them using evolving best practices. Key components of this strategy include:
- Regular review cycles: Laws should include built-in mechanisms for periodic updates to ensure they remain relevant as technology advances.
- Performance-based standards: Focus on measurable outcomes like accuracy, fairness, and security rather than prescribing specific technical architectures.
- Industry-government collaboration: Create formal channels for researchers, developers, and policymakers to share insights and address emerging risks in real time.
- Transparency requirements: Mandate clear documentation of model capabilities, limitations, and intended use cases to help users and regulators make informed decisions.
This flexible approach acknowledges that AI is not a static product but a continuously evolving field. Regulations that can bend without breaking are far more likely to succeed in the long run.
What This Means for the Future of AI Governance
The conversation around AI regulation is no longer about whether we need rules, but how we design them to last. Companies that prioritize safety and transparency are increasingly stepping up to help shape these frameworks, recognizing that responsible governance ultimately benefits the entire industry. Clear rules reduce legal uncertainty, build public trust, and create a level playing field for innovation. State governments that embrace adaptive, collaborative approaches will likely emerge as leaders in shaping the next generation of AI policy.
As artificial intelligence continues to weave itself into the fabric of modern life, the margin for error shrinks. The push from companies like Anthropic highlights a growing consensus: regulation cannot be an afterthought. It must be proactive, agile, and grounded in a realistic understanding of how technology actually develops. By urging states to move faster and build smarter frameworks, the industry is signaling that it’s ready to partner with policymakers in creating a system that protects users while preserving the room needed for innovation to thrive. The race to regulate AI is far from over, but with the right approach, it doesn’t have to be a race against time.
