A faster and fairer ADF through AI

When pundits consider the application of AI and machine learning in a military context, most immediately consider programs such as the ‘Scarlet Dragon’ AI platform.

The concept of Scarlet Dragon is simple. It scans satellite imagery and uses machine learning to identify potential targets which are then relayed to a human operator to oversee fire control orders. Throughout the most recent iteration of Scarlet Dragon testing, the platform analysed 7,200 square kilometres of satellite radar imagery, which was then notionally relayed to F-35s, F-15s, and F-18s for fire support.

Others think of Project Maven – a co-operative machine learning leap between some of the world’s largest tech companies and the Pentagon. Project Maven builds the target identification algorithms to ensure that the weapons platform can correctly identify friend from foe, military vehicle versus civilian car. In fact, Google supported the Pentagon’s Project Maven until a mass employee petition in 2018 forced Google bosses to withdraw.

Beyond the glamours combat applications of AI, AI and machine learning can also be used to automate the most basic functions of a military.

Writing in the Australian Army Research Centre Journal, Captain Samuel White analysed how machine learning can streamline the decision making processes in legal cases, administrative action and career management procedures – freeing up capabilities to help build a stronger and more competitive Australian Army.  

“The day-to-day administration of personnel constitutes a significant burden on any commander, detracting from the ability to conduct training to prepare for combat,” CAPT White noted. “This paper suggests a possible method for the Australian Defence Force (ADF) to help reduce the cognitive clutter surrounding its administration, policy and military discipline through the use of machine learning algorithms.”

Automating administration and policy via machine learning isn’t a new phenomenon, CAPT White argues. Such processes have already been undertaken in several government initiatives including the tax system, migration and social security. For his theory, CAPT White relies on machine learning as a “subset” of AI.

VIEW ALL

“ML as a subset of AI uses statistical methods to enable computers to improve with experience using non-linear processing. It has shown itself useful for particular tasks and activities such as sorting data, finding patterns and trends, and completing a high-volume of repetitive tasks quickly while minimising errors,” CAPT White argued in his journal entry.

Simply, machine learning would analyse all previous data relating to the ADF’s cases and provide advice or recommendations to the Commander for possible actions.

This machine learning capability could have a profound impact on the application of disciplinary action in the ADF, not least because “a majority” of legal disciplinary action is conducted by military commanders who do not have a background in law, CAPT White notes. Several other benefits include consistent application of the law regardless of all qualitative factors influencing the proceedings and that the algorithm itself would improve when paired with more datasets.  

However, CAPT White flagged that the algorithms could be liable to bias in the datasets, generating skewed recommendations.

“A review of the publicly available data at the time of publication, with respect to superior tribunals reveals that junior soldiers are charged, and found guilty, twice as often as non-commissioned officers (NCOs),” CAPT White noted.

“Yet this data, if fed improperly to an algorithm, would suppose that soldiers were statistically more likely to offend than NCOs or officers, and could potentially sentence them in a more severe manner.”

Despite these drawbacks, CAPT White’s analysis continues by discussing the beneficial impacts of machine learning to streamline the administrative action process.  

It is common knowledge that military doctrines and pams are behemoth documents. With non-legally trained Commanders often overseeing the application of such manuals, it is an understandable – and perhaps unavoidable – circumstance when Army guidelines are misapplied in certain contexts.

As such, CAPT White notes that machine learning can produce relevant facts or policy to guide and inform the Commander throughout their decision-making process in situations of administrative action.

“Take, for example, the Military Personnel Policy Manual (MILPERSMAN), the unclassified public copy of which is 745 pages. MILPERSMAN Part 4, Chapter 1 relates to the use and abuse of alcohol. This policy has differing thresholds across the three Services as to what administrative sanctions may, or must, be initiated on the basis of the number of alcohol incidents or the alcohol blood level,” CAPT White notes.

Finally, CAPT White argues that machine learning would also support career management in the military. The creation of an online database for Commanders to input real-time information using a series of key-words would enable the machine learning algorithm to produce greater insight into the member’s capability relative to their peers.

“If the appraisal is done month to month, the ML algorithm can moderate the individual across the years. In the annualised report, if a person has a poor few months before their appraisal, the supervisor may be biased by that poor performance instead of accurately weighing all the information recorded,” CAPT White noted.

The strength of the tool relies in averaging out the Commander’s biases for or against specific individuals and providing a fairer ranking system relative to those with whom they operate.

While AI and machine learning is still a developing field, there are green shoots as to its application not only in wartime but to streamline the army itself. Minimising the decision-making process for Commanders will free their time, enabling them to return to getting the Australian Army war ready.

Get involved with the discussion and let us know your thoughts on Australia’s future role and position in the Indo-Pacific region and what you would like to see from Australia's political leaders in terms of partisan and bipartisan agenda setting in the comments section below, or get in touch with This email address is being protected from spambots. You need JavaScript enabled to view it., This email address is being protected from spambots. You need JavaScript enabled to view it., or at This email address is being protected from spambots. You need JavaScript enabled to view it..