Many Questions Remain Unanswered About Big Tech, Defense Use of AI

January 25, 2021—The need for clear guidelines facilitating the development and use of artificial intelligence continues to cause raging debates among those working in the public sector. While private tech companies, like Google and Microsoft, have already created their own ethical guidelines, layin

Many Questions Remain Unanswered About Big Tech, Defense Use of AI
Photo of Alka Patel, Head of AI Ethics Policy at the Department of Defense’s JAIC, by CMU

January 25, 2021—The need for clear guidelines facilitating the development and use of artificial intelligence continues to cause raging debates among those working in the public sector.

While private tech companies, like Google and Microsoft, have already created their own ethical guidelines, laying out how the companies will use AI software, the U.S. Department of Defense is just beginning to tackle the process of outlining procedures that detail the government’s use of AI.

Creating resources that detail how AI technology functions to assist the DoD, and other federal agencies is key in leveling the AI development gap, which persists between private companies and government entities, according to a panel of federal employees who convened for a Nextgov webinar entitled Genius Machines on Thursday.

When it comes to government use of AI, there are many questions that remain to be answered. Since the technology holds such a variety of capabilities, panelists agreed that there need to be conversations held surrounding ethical requirements, policies for specific use cases, regulations for software development, and more.

However, with the creation of the Joint Artificial Intelligence Center, the DoD is making moves to centralize AI efforts across the government. Federal agencies are also focusing on incorporating these tools to detect insider threats, increase efficiency and maintain a tactical edge on the battlefield.

The JAIC works to maintain good relationships between the Pentagon and Silicon Valley, and further, to help federal agencies understand the power and responsibility that comes along with producing AI.

Alka Patel, head of AI Ethics Policy at the DoD’s JAIC, said that regardless of the use case, there must be specific assessments and considerations taken into account for all government-use of AI.

Patel maintained that a governance structure with guidelines and principles must be designed, saying there is a need for the DoD to work towards standardizing best practices and risk management assessment tools. Once these basic guidelines are built, Patel believes using AI tools will become second nature.

Pentagon leaders have been clear, if their efforts to leverage emerging technology are to show real value, the Defense Department must tap the innovation coming out of the private sector. Patel said that in order to help federal agencies design better products, the public and private sectors need to collaborate.

Through collaboration, companies are able to engage with the DoD, to ensure their technology can fulfill what the federal government requires. Creating more avenues to engage with all stakeholders is a critical point to leveraging AI to its greatest potential use.

Such engagement with acquisition partners creates a conversation between all parties to work together. Developing technologies is in the best interest of all, said Patel.