Resilience in the age of AI

The OpenAI Foundation exists to ensure that artificial general intelligence benefits all of humanity.

We have been working relentlessly to keep pace with the rapid advances in AI itself. In April, the Foundation announced⁠ our first $100 million of grants in Life Sciences and Curing Diseases, with the ambition to help prevent and treat diseases like Alzheimer’s by harnessing advanced AI. Last week, we announced⁠ our Jobs and Economic Futures program, with the hope of understanding and shaping what work and economic prosperity mean to future generations.

Today, we are expanding on our vision for the next major program—ensuring that, as AI’s capabilities grow, society’s ability to harness it will grow just as fast. We call this work AI resilience: the ecosystem approach required to mitigate the risks of AI, so that society can maximize its benefits.

Our work has already begun. In the few short months since initiating our work, the Foundation is working to finalize more than $130 million in grants to organizations through our AI Resilience program, to be shared publicly soon and with more to come.^1⁠

The pattern of transformational technologies

The importance of AI resilience is best understood through the lens of past technologies that meaningfully shaped human history.

Every so often, a technology arrives that reshapes society from the ground up. Economists call these “general purpose technologies.” Fire. The printing press. Electricity. The internet. Each followed a similar arc: rapid innovation, real risks, and institutions racing to catch up. But each example also shows what it takes to make a powerful technology safe.

Fire made human civilization possible. It kept us warm, cooked our food, and protected us from predators. It also burned our cities to the ground. Over time, societies built resilience: fire-resistant materials, hydrant networks, professional fire departments, and building codes. An ecosystem, layer by layer.

Electricity followed the same path. After Edison’s Pearl Street Station lit up Manhattan in 1882, electricity brought fires, electrocutions, and public panic. Without protections like insulated wires, breakers, and codes, workers and bystanders were electrocuted in cities across the country. Cities debated whether the experiment should be abandoned entirely. Instead, as the technology advanced, we set up independent testing bodies like Underwriters Laboratories, industry standards like the National Electrical Code, and public investment that brought power to communities the market had left behind. Each layer made electricity safer and more accessible; today, it is so safe that a child can flip a switch and light appears.

This is what resilience looks like when it’s done well.

AI requires a resilience ecosystem

AI is following the same trajectory as previous technologies, but moving at an unprecedented speed.

We are still in its early days, but the benefits are already clear: AI is lowering the barriers to starting a business, expanding access to education, accelerating scientific discovery, and transforming medicine.

At the same time, the risks are emerging just as quickly—and in a mirror image of AI’s benefits. The same growth that creates new industries can upend existing ones and disrupt careers. The same systems that can help young people learn and create could also lead to adverse behavior. The tools that speed biological research could lower the barrier to creating harmful pathogens. And AI’s ability to write code, in the wrong hands, could threaten critical infrastructure.

The early OpenAI team believed that ensuring AI benefits society depended primarily on solving the technical alignment problem. That remains critical—and central to our work—but we now believe it is only one part of the puzzle. As AI diffuses across sectors and nations, society will also require independent research, public infrastructure, industry coordination, and entirely new fields of expertise. In short, it will require AI resilience.

We have chosen to focus our initial work in four areas^2⁠ that sit at the intersection of large, near-term risks and immediate impact:

1. Bio-resilience to help prevent engineered pandemics of the future;

2. Cyber-resilience to work to ensure the safety of our world’s critical systems;

3. AI model safety to solidify humanity’s control of the models we create; and

4. AI’s impact on young people to help make technology a positive force for future generations.

Our work is just beginning. We plan to share more about our strategies and initial grants in each area, and to expand to other areas over time.

Bio-resilience

AI will enable biological research to move at unprecedented speed, helping develop new cures and public health improvements that enable us to all live healthier and longer. However, these same capabilities could also be misused by malicious actors, lowering the barrier to designing harmful pathogens.

The age of AI requires a renewed focus on biosecurity. Because advanced AI systems could be misused by bad actors to help create a wide range of biological threats, we will prioritize pathogen-agnostic biosecurity solutions. This will require investments across prevention, detection, and defense. We need to make it harder for malicious actors to access the expertise, equipment, and materials to create biological threats, improve our ability to identify and track novel outbreaks early, and strengthen the technologies—such as protective equipment, indoor air cleaning systems, and medical countermeasures—needed to respond quickly and effectively.

Cyber-resilience

AI has begun to rapidly reshape the cybersecurity landscape. The work that once required specialized teams can now be assisted or automated by capable models. At the same time, rapidly-improving AI capabilities can also be used to accelerate cyberdefenders, including by identifying and patching vulnerabilities and accelerating response.

Many large companies and private actors can spend heavily on cyber to secure their own systems, including with new advances in AI. We anticipate focusing significant resources on securing other important societal actors that are less resourced and will have a much harder time deploying AI-ready cyberdefenses as quickly as needed. In parallel, we are also focused on preparing for novel security challenges that artificial general intelligence will ultimately bring.

AI model safety

AI model safety focuses on the behavior of the systems themselves—whether they are truthful, reliable, and aligned with human intent. In a world where this goes awry, models can break out and behave in unpredictable ways, deceiving us or pursuing goals beyond their design. Getting this right becomes increasingly important as AI systems grow more autonomous and approach—and eventually surpass—human-level intelligence.

AI companies are investing substantial resources in model safety. However, the importance of this challenge calls for a broader, more robust ecosystem: independent institutions to evaluate model safety, public infrastructure to verify models’ safe deployment in practice, and continued advances in alignment science that advance the field broadly.

AI’s impact on young people

Young people are often the earliest adopters of new technologies, using them to learn, create, communicate, and explore the world. AI is no exception. But as these tools become an increasing part of young people’s daily lives, it is critical that we develop a stronger evidence base to understand its impacts.

Families, schools, policymakers, and community organizations are all grappling with questions about how and when young people engage with AI—including its impact on human connection, learning, and development. Our initial focus will be on advancing independent research to help guide those decisions—to better understand where AI can support development, the risks it may introduce, and the contexts that shape those effects.

These insights should drive broad safety standards and design principles that guide how any AI product is developed, how schools choose to deploy them, and if and how families decide to incorporate these technologies into their lives.

The work ahead

There is one critical difference between AI and the technologies that came before it: speed.

Fire resilience took millennia. Electricity resilience took decades. AI resilience is evolving in a matter of years. The systems that make it safe, reliable, and broadly beneficial must be built alongside it.

If we get it right, AI can become part of the foundational infrastructure of modern life—expanding access to knowledge, accelerating discovery, and improving lives at a global scale.

But that outcome isn’t guaranteed. No general purpose technology ever made itself safe.

Resilience is a permanent discipline that requires many people and institutions to build, invest, and collaborate. That is the work ahead, and it is one of the defining challenges of our time. We hope you’ll join us.