AI's Potential for Autonomous Warfare

Artificial intelligence (AI) is a rapidly advancing field with implications that reach far and wide. One of the areas where AI's potential impact is causing significant concern is in military applications.

In September last year, the US Air Force conducted its first dogfight between a human pilot and an AI-controlled fighter jet at Edwards Air Force Base in California. The aircraft involved were both F-16 jets, one manned and the other controlled by a computer.

The planes engaged in offensive and defensive manoeuvres as well as air combat within visual range while reaching top speeds of 1200 mph. Footage published by the US Defense Advanced Research Projects Agency (DARPA) showed them swooping around each other at high speed across clear blue skies

While no official information was released about who won this aerial duel, it should be noted that an AI had already defeated a human operator 5-0 in simulated dogfights back in August 2020. This fact marks what DARPA described as "a transformational moment" for aerospace history: machine learning has now been successfully used to control autonomous planes.

This advancement implies substantial changes on future battlefields, including those involving peacekeeping forces. Autonomous drones capable of hunting down targets without any human input are currently being developed around the world. These machines have raised alarm bells among many observers due to their lethal capabilities.

A report from last year cited by The New York Times revealed how, during Libya's civil war, government-backed forces stationed in Tripoli used what was referred to as "a lethal autonomous weapons system.". This drone reportedly attacked enemy soldiers fleeing rocket assaults without any direct command from humans.

Such incidents highlight why there's growing concern over autonomous weapons systems: they remove ethical considerations usually associated with warfare decisions made by humans. Who will bear responsibility if such machines cause civilian casualties or target hospitals, journalists, or humanitarian workers?

Moreover, algorithms controlling these systems can't predict all possible outcomes accurately. There's a risk they might misinterpret situations and unintentionally escalate conflicts.

The United Nations (UN) has already expressed its concern about this issue, passing a resolution stating that an algorithm should never have complete control over decisions involving the killing or harming of humans. It also stressed preserving human responsibility and accountability in any use of lethal force, regardless of the type of weapon system involved.

The UN's resolution also highlighted concerns about potential negative impacts on global security from autonomous weapons systems. These include risks such as lowering conflict thresholds, triggering arms races, and making it easier for non-state actors to access deadly technologies.

Given these challenges, it's imperative for the international community to address this pressing issue urgently by seeking input from member states, observer states, and various stakeholders, including civil society.

As AI continues to evolve rapidly across industries worldwide, Fiji cannot afford to be left behind in understanding its implications—both positive and adverse—for our society. We need a comprehensive understanding of AI applications so we can leverage them effectively while mitigating associated risks.

In conclusion, while artificial intelligence holds tremendous promise for technological advancements across many fields—including military applications—we must remain vigilant against the potential dangers inherent in new technologies.