News

2021-08-01

How Can AI be Used Ethically in Defence?

A report by the U.S. National Security Commission on Artificial Intelligence (NSCAI) highlights emerging consensus on the principles for using AI ethically and responsibly for defence and intelligence applications.

As the report states, if artificial intelligence (AI) systems do not work as designed or are unpredictable, “leaders will not adopt them, operators will not use them, U.S. Congress will not fund them, and the American people will not support them.”

That is why justified confidence is so important for AI-enabled systems. Justified confidence in AI is about developing systems that are robust, reliable, and accountable, and ensuring these attributes can be verified and validated.

AI is already ubiquitous in everyday lives, from streaming services to navigation apps to secure banking. But it is also playing a role in defence, such as way-finding for unmanned vehicles, automated target recognition, and many other applications that focus on speed, scale and efficiency.

The power of AI is its ability to learn and adapt to changing situations. The battlefield is a dynamic environment and the side that adapts fastest gains the advantage.

But like with all systems, AI is vulnerable to attack and failure. To truly harness the power of AI technology, developers must align with the ethical principles adopted by the U.S. Department of Defense. However, delivering on the promise of robust, reliable, and accountable AI systems requires a team effort – industry, government and academia all have roles to play.

Integrated Approach

To achieve this, companies such as Northrop Grumman are working on developing a cohesive policy and governance processes for AI, spanning from development to testing and operations.

One of Northrop Grumman’s partners are a Silicon Valley startup, Credo AI. They are sharing their governance tools to apply comprehensive, relevant ethical AI policies to guide AI development. The company is also working with universities like Carnegie Mellon to develop new secure and ethical AI best practices, in addition to collaborating with leading commercial companies to advance AI technology.

Another step the company is taking is to extend its DevSecOps process to automate and document best practices in the development, testing, deployment, and monitoring of AI software systems.

Critical to success is Northrop Grumman’s AI workforce. The company continues to invest in a mission-focused AI workforce through formal training, mentoring and apprenticeship programmes.

According to a company statement, Northrop Grumman strives for equitable algorithms and minimises the potential for unintended bias by leveraging a diverse engineering team and testing for data bias using commercial best practices, among other monitoring techniques.

The company has developed tools to provide a log of data provenance, ensuring traceable, transparent, and auditable development processes. Leveraging best practices, its work in AI governance enables robust risk assessment, algorithmic transparency and graceful termination when required.

This integrated approach from development to operation is essential to achieving justified confidence in AI-enabled systems.

Comments () Views (5547 ) Print

Send To Friend