Why OpenAI has hired ex-US Army secretary Dan Driscoll for 'security' role

The appointment comes as OpenAI steps up its work with defence agencies and governments on AI applications in national security and cybersecurity

OpenAI has brought in a former secretary of the US Army to help it figure out how to put artificial intelligence to work in national security, from improving military operations to strengthening cyber defences.

 

The company on Monday announced the appointment of Dan Driscoll, who served as the 26th Secretary of the US Army, as its senior counsellor for national security modernisation. The newly created role will involve advising OpenAI's leadership and its public- and private-sector partners on how advanced AI can be used to strengthen national and cybersecurity.

 

In its announcement, OpenAI said Driscoll would work with governments, industry, academic institutions and US allies to identify practical ways of using AI in national security. He will also help the company's teams understand the operational requirements of security institutions around the world.

 

Who is Dan Driscoll?

Driscoll previously served as the 26th Secretary of the US Army, overseeing its operations, modernisation and resources. His responsibilities covered nearly one million active-duty, National Guard and Reserve soldiers, as well as more than 265,000 Army civilian employees, according to OpenAI.

 

During his tenure, he worked on accelerating the adoption of new technology, reforming military procurement and modernising the Army.

 

He has also served as acting director of the US Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF).

Why has OpenAI hired a former Army secretary?

Driscoll's experience in the military and government is expected to help OpenAI understand what defence institutions need from AI tools and how these can be put to practical use.

 

In its announcement, the company said he would advise public- and private-sector partners on changes to policy and strategy that could help unlock AI applications in national security. His role will also involve identifying areas where military requirements, institutional changes and emerging AI capabilities intersect.

 

Driscoll said military personnel needed better tools to keep pace with rapidly changing national security requirements.

 

"The people who serve deserve the very best tools we can put in their hands," he said, adding that AI could help them cut through paperwork, make sense of larger volumes of information and get the right answers to the right people faster.

 

He said he was joining OpenAI to help the US and its allies use the technology responsibly.

 

OpenAI CEO Sam Altman said the US had an opportunity to work with other countries to expand the adoption of what he described as "democratic AI" and strengthen American leadership.

What is OpenAI's Daybreak programme?

The appointment also builds on OpenAI's Daybreak programme, through which the company is working with allies and partners to strengthen cyber resilience and protect essential services.

 

The initiative focuses on helping organisations identify and address software vulnerabilities before they can be exploited. This is particularly relevant for critical infrastructure and institutions that may face cyber threats.

 

OpenAI's work in this area has included supporting cyber-defence efforts in European countries, including France, Germany and Poland. It has also provided Ukraine access to tools to help identify software vulnerabilities, according to the Financial Times.

OpenAI's growing defence ambitions

The hiring comes as OpenAI expands its relationship with the US government and defence establishment.

 

Earlier this year, the company agreed to deploy its AI models in classified settings for the US Department of Defence. The development followed the collapse of negotiations between the Pentagon and rival AI company Anthropic over the terms governing military use of its technology.

 

The use of AI in defence has raised questions about how the technology can be deployed safely, particularly in sensitive military and intelligence operations. While AI can help process information and identify security threats, its use also requires safeguards and human oversight.