OpenAI public policy agenda

OpenAI’s mission is to ensure that artificial general intelligence (AGI) benefits all of humanity. Our work is guided by five core principles⁠ that shape how we build AI and how we engage in public policy: 

1. Democratization. We will resist the potential of this technology to consolidate power in the hands of the few.
2. Empowerment. We believe AI can empower everyone to achieve their goals, learn more, be happier and more fulfilled, and pursue their dream, and that society as a whole will benefit from this.
3. Universal prosperity. We want a future where everyone can have an excellent life.
4. Resilience. AI will introduce new risks, and we will work with other companies, ecosystems, governments, and society to solve them. 
5. Adaptability. We continue to believe the only way to meet the challenges of a very unpredictable future is to be prepared to update our positions as we learn more. 

We believe AI has the potential to reshape how people work, learn, create, and participate in society, and that democracies will play an essential role in helping broaden access to opportunity, mitigate risks, and ensure people have real agency to shape the AI future they want. That’s why we’re committed to making our technology freely and safely accessible. 

Our user base reflects that commitment:⁠ we have as many women as men using our tools, more users both under and over age 30 than any other AI platform, and more users earning under $100,000 than over $100,000—mirroring the broader global workforce.

That’s why OpenAI engages with governments and civil society around the world: to help policymakers understand our technology and its impacts, improve how governments function and deliver value to their constituents, expand the ability of people to participate in and benefit from democratic institutions, and shape policies that advance our mission of ensuring that AGI benefits all of humanity.

The following priorities reflect how we translate our mission⁠ and our principles⁠ into public policy. We support policies that help governments and society respond to the opportunities and challenges created by AI, mitigate risks, expand access to opportunity, and ensure people can meaningfully participate in and benefit from the AI economy. Together, these priorities are intended to help governments better serve their constituents while advancing our mission of ensuring that AGI benefits all of humanity. They represent some of the areas where we are most actively engaging today, but this list is not exhaustive. As AI evolves, we also expect our policy priorities and areas of engagement to continue to evolve as well.

We believe frontier AI safety is a national security and public safety issue, particularly for the most advanced general-purpose AI models, which can create risks related to cyber, chemical, biological, radiological, and nuclear (CBRN) weapons. In the US, we support state efforts to align around common frameworks such as California SB 53⁠(opens in a new window),the New York RAISE Act⁠(opens in a new window), and Illinois SB 315⁠(opens in a new window) which emphasize transparency, public reporting around catastrophic-risk evaluations and safety incidents, whistleblower protections, and enforceable accountability for developers that fail to meet their safety and security responsibilities. These state-level approaches can help establish harmonized standards that reduce fragmentation and create a path toward an eventual federal framework. 

We also support Congressional action to establish a comprehensive federal framework that leverages the emerging consensus reflected in state frontier safety laws; strengthens the Center for AI Standards and Innovation (CAISI) as the US federal government's primary institution for frontier AI safety; and mobilizes a broader resilience plan across government to address the national security and public safety challenges posed by frontier AI. This framework should require CAISI to conduct evaluations of the most capable frontier models, direct CAISI to create an independent assessment ecosystem, and prioritize monitoring progress towards recursive self improvement (RSI). With a comprehensive federal framework in place, we support the preemption of state laws that seek to regulate the same frontier safety risks.

We also support the US federal government playing a leading role setting common international standards, and were the first US company to sign the EU AI Act Code of Practice. We were also among the first companies to enter into voluntary agreements with both the US CAISI and the UK AI Security Institute (UK AISI). 

Looking to the future, we believe policymaker focus should begin considering more ambitious ideas⁠(opens in a new window) such as model containment playbooks, incident reporting, or international governance bodies that facilitate coordination around frontier AI risks, safeguards, and security incidents.

On cybersecurity⁠(opens in a new window) in particular, we support policies that expand trusted access to AI-powered cyber defense tools and stronger partnerships among governments, researchers, and industry to conduct evaluations, deepen information-sharing, and build resilience measures to strengthen cyber defense. We also support efforts to modernize outdated public-sector cybersecurity systems and are partnering with the U.S. government at the federal, state, and local levels, and with international partners as well.

We believe AI can help young people learn, create, and develop the skills and entrepreneurial mindset they will need to thrive in the future economy—provided it is deployed with strong safeguards and age-appropriate protections. Teens should have access to safe and trustworthy AI⁠(opens in a new window) at home, at school, and as they prepare to join the workforce, and they should be protected from its potential harms. We support strong, enforceable, risk-based regulations for teen safety and frameworks that pair robust safeguards with transparency and tools that put families in control. 

That includes requirements for privacy-preserving age assurance so companies can identify when a user is a minor  and apply age-appropriate protections; regular youth safety risk assessments that identify foreseeable risks and drive proportionate safeguards before harm occurs; accessible parental controls that help families guide and support their teen’s experiences; and clear public youth  safety policies that explain what protections are in place and how they evolve over time.

It also includes safeguards designed to address the risks teens may encounter when interacting with AI systems, including protections against harmful or age-inappropriate content, clear protocols for responding to serious safety situations, and safeguards against manipulative, or deceptive interactions. Crisis-response protocols should include in-service support, referrals to appropriate real-world resources such as 988 in the United States, and timely parental notifications when appropriate. Companies should also be required to protect teen's personal information, including by restricting targeted advertising to minors and the sale of personal data.

Finally, strong accountability mechanisms, including independent audits, are essential to ensure these protections are meaningful in practice. Audits should be underpinned by a set of common standards that allow for interoperable audits across jurisdictions. Legislative frameworks should include oversight and enforcement measures that allow governments to assess whether companies are effectively implementing youth safety safeguards, mitigating identified risks, and complying with youth safety and privacy obligations. We are proud to support the Parents and Kids Safe AI Coalition’s⁠(opens in a new window) efforts to advance these principles and promote practical AI safeguards that protect young people while preserving their ability to learn from and benefit from AI. 

Additionally, we support strong legal and technical protections against child sexual abuse and exploitation. This includes modernizing child protection laws to address the ways generative AI can be misused to create synthetic CSAM, digitally alter existing imagery, and scale grooming or exploitation activity. Policymakers should ensure CSAM statutes clearly cover AI-generated and digitally altered material, clarifying liability for intentional efforts to produce or solicit CSAM, and preserving strong enforcement authority for prosecutors and law enforcement. We also support provider reporting and coordination standards that improve the quality and actionability of CyberTipline reports, while reducing investigative burden and helping NCMEC and law enforcement act faster. Finally, companies should implement safety-by-design safeguards—including detection, refusal mechanisms, human oversight, and continuous monitoring—to interrupt exploitation attempts before harm occurs.

AI will increasingly shape how people learn, work, and participate in civic life. Educational institutions will play a critical role in building AI literacy⁠(opens in a new window) and preparing students to navigate a world where AI is ubiquitous. We support policies that help students, teachers, families, and communities engage with AI safely, critically, and creatively, while ensuring educators remain central to classroom decision-making and set the pace for how AI is adopted in schools. That includes investments in AI literacy, strong core instruction in subjects like history, civics, math, science, literature, computer science, and career-technical education. It also includes teacher training and protected time for professional learning, workforce-aligned learning pathways, and expanded access to AI tools, broadband, devices, and educational resources for schools, libraries, and community institutions. We also support efforts to strengthen research into how AI impacts learning outcomes, student wellbeing, and educational equity and efforts to incorporate learnings into AI development and deployment.

We believe that everyone should have the ability to participate in the new opportunities that AI creates⁠(opens in a new window), and that’s why we make ChatGPT available for free. We support policies that expand affordable access to useful AI and help workers, entrepreneurs, educators, and small businesses adopt it through investments in workforce training and AI literacy. We also support the creation of regional AI hubs that connect employers, labor organizations, community colleges, universities, workforce boards, and local businesses, and small-business adoption programs that provide access to AI tools, technical support, and hands-on training.

We also partner with labor organizations⁠(opens in a new window) to expand access to training⁠ to help workers build practical AI skills and prepare for changes in the economy driven by increasingly capable AI systems, and we regularly publish data on how workers and employers are using our tools and we support workforce transparency measures that not only help policymakers and the public better understand how AI is reshaping work but also help identify pathways that help workers move into new roles and human-centered work. At the same time, we are committed to working with policymakers to develop and implement a more ambitious workforce and economic transition agenda⁠(opens in a new window), which contemplates portable benefits, tax modernization, public wealth funds, and adaptive safety nets.

We believe AI can expand creative expression, empower artists and creators, enable new forms of expression, commentary, and creative transformation, and lower barriers for independent creators, small studios, and new voices to participate in the creative economy. At the same time, people should be able to understand where digital content came from and be protected from harmful deepfakes such as deceptive impersonation and unauthorized digital replicas. We support policies that promote transparency online, including requirements for AI tools to include provenance signals, such as those developed by the Coalition for Content Provenance and Authenticity (C2PA), in the audiovisual content they generate, and are working with partners across industry to advance interoperable provenance standards and content authenticity tools that help people better understand where digital content came from. We support policies that protect against the harmful misuse of voice and likeness while preserving important safeguards for free expression, parody, journalism, and other lawful uses. 

We support strong protections against image-based sexual abuse, including efforts to criminalize the distribution of synthetic non-consensual intimate imagery and strengthen remedies for victims. We also support election integrity protections related to deceptive AI-generated political content, including prohibitions on intentionally distributing AI-generated content to mislead voters, as well as disclosure requirements for certain AI-generated political advertising and campaign communications. We partner with trusted election authorities to help provide accurate voting and election information⁠ and to help prevent our systems from unduly recommending or ranking candidates.

We believe that if AI infrastructure is developed in partnership with communities, it is an engine of local, state, and national economic growth. Partnering with local communities starts with understanding local needs, and that’s why every Stargate site has a tailored community plan⁠ developed in consultation with communities including on site design and energy sources. It’s also why we support policies that require transparency around water, electricity, and government agreements, with appropriate exceptions to maintain security, commercial sensitivities, trade secrets, and proprietary information. 

We commit to paying our own way on energy, so that our operations don’t increase electricity prices and we also support policies that ensure datacenters pay their fair share of electricity costs, including requirements for state public utility commissions to create large load tariffs that require datacenters to pay the incremental costs they create. We are working with local utilities and regulators to ensure datacenters help contribute to grid reliability. We also support sustainability reporting requirements, including disclosures about technologies and practices that reduce environmental impacts, such as low-emission backup generators and closed-loop or low-water cooling systems. We support policies that help local workers access good paying construction jobs, including incentives to use registered apprentices and project labor agreements. And we encourage policies that maintain industry-neutral and predictable tax and incentive frameworks, hold companies accountable for delivering on their commitments, and catalyze innovation, entrepreneurship, and long-term economic growth in surrounding communities.