When I served as White House Chief Information Officer under President George W. Bush, my responsibility was protecting the technology the federal government relied on every day while anticipating future threats. That same mindset is urgently needed as America enters the next era of artificial intelligence.
President Donald Trump has an opportunity very few presidents have ever received: the chance to shape the future of a technology that will transform our economy, national security, military capabilities, and everyday lives. The decisions made during this administration will determine whether America continues leading the AI revolution or allows adversaries like China to set the rules of the next technological era.
The recent dispute between the White House and Anthropic over its newest frontier AI models should serve as a wake-up call. After officials raised concerns about a potential security vulnerability known in the industry as a “jailbreak,” the administration moved to restrict international access to Anthropic’s latest models.
The company and government officials reportedly disagreed over the severity of the risk. Anthropic argued the vulnerability was limited, while federal officials viewed the potential national security implications differently.
The lesson is bigger than one company or one model. Technology has advanced faster than the systems we have built to secure it, and America cannot afford to create artificial intelligence policy one emergency at a time.
President Trump understands something Washington too often forgets. American strength comes from American innovation. We cannot regulate ourselves into second place, but we also cannot ignore the reality that powerful technologies require serious safeguards.
The first wave of generative AI introduced millions of people to tools capable of answering questions, drafting documents, and creating images. The next generation is far more powerful — as frontier AI models, Loop engineering, and Agentic AI become capable of stronger autonomy, reasoning, coding, and processing enormous amounts of information.
Agentic AI represents the next major shift, moving from tools that simply respond to humans toward systems that can independently complete tasks. Instead of asking a chatbot a question, a person may soon give an AI system a larger goal. The AI can create a plan, interact with other software, analyze results, and execute complex workflows with limited human involvement. The AI could create its own AI agent helpers and teach itself new skills without a specific human prompt.
The opportunities are enormous, but so are the risks. As someone who has spent my career focused on implementing cutting-edge technology, innovation, and cybersecurity, I know an important reality that we typically do not design with the human in mind, and that no technology system is perfectly secure, including artificial intelligence.
The goal cannot be building an AI model that is impossible to manipulate. Every major technological advancement brings vulnerabilities that must be discovered, measured, and addressed. The better question is how we determine which vulnerabilities represent manageable risks and which create true national security threats.
That is why the reported discussions between the White House and Anthropic to create shared benchmarks for evaluating AI security flaws are an important step forward. Cybersecurity professionals have long understood that not every vulnerability is the same. A minor weakness, a theoretical exploit, and a catastrophic breach require very different responses. Artificial intelligence needs that same level of maturity.
The government has a responsibility to prevent our adversaries from exploiting advanced technology. Frontier AI could reshape cybersecurity, intelligence operations, economic competition, and military capabilities, which makes protecting America’s advantage one of the most important national security challenges of our time.
President Trump has the opportunity to build the foundation for a new era of American technological dominance that protects our country while unleashing the entrepreneurs, engineers, and companies that make the United States the greatest innovation engine in the world.
However, how the government responds matters. Businesses, innovators, and our allies need clear rules because sudden interventions create uncertainty and make it harder for responsible companies to plan, invest, and compete.
The nations that control the most advanced AI infrastructure will have significant strategic advantages, and the countries that get this right will define the next century. The private sector is already adapting as companies seek greater flexibility and control over their AI systems through lower-cost models, open-source alternatives, and multi-model approaches.
America does not need an AI strategy built around reacting to crises. We need one designed before the crisis happens, with clear technical standards, transparent security evaluations, strong enterprise protections, audit trails, responsible access controls, and meaningful cooperation between government and industry.
It also means remembering that artificial intelligence remains a human responsibility. As Pope Leo XIV warned world leaders during discussions about AI, humanity cannot surrender its own decision-making authority to machines. Technology should strengthen human judgment, not replace it.
Artificial intelligence is no longer just another software tool. It is rapidly becoming a critical infrastructure that will shape businesses, governments, and national security for generations.
America has led every major technological revolution because we have embraced innovation, competed aggressively, and led with confidence. President Trump has a chance to make American AI leadership a defining legacy of his presidency if we build the strategy now, rather than waiting until the next crisis forces us to act.
***
Theresa Payton served as White House Chief Information Officer under President George W. Bush and was the first woman to hold the position. She is currently the CEO of Fortalice Solutions, a cybersecurity and intelligence operations company.
.png)
.png)
