Skip to main content

Nights on the vehicle checkpoint outside the regional hospital were turning into a real drag. The city had been liberated two months ago and the insurgency driven out. The majority of attacks were now skirmishes on the outskirts of town. The key roads leading into the city had been destroyed by the orbital kinetic bombardment system, leaving massive craters where key roads had once existed. Now there were only three heavily guarded choke points accessible to vehicle traffic which the enemy would need to capture if they had any hope of taking the city.

Displaced persons still streamed through the cratered mess, many with injury and illness from their long journey across the war-torn country, so the hospital was operating at over capacity. So whilst there was a steady stream of comings and goings the people seeking the services of the hospital were relived for the relative safety within the city and content to wait as their needs were triaged and they were told to wait until they could be seen to.

With the hospital separated from the closest access road by nearly 50km of sprawling urban slum there was little chance of anything eventful occurring tonight, or any of the following nights on CPL Adam Jacobson’s deployment. With the large number of people being forced to enter the city on foot vehicle traffic was also minimal. The few vehicles his team did have to search were usually trucks delivering crucial supplies. There was sometimes the odd private vehicle. An NGO had acquired a small fleet of self-drive capable vehicles and were using them to shuttle to most critical refugees to the hospital, but with the trip to and from taking longer by vehicle than on foot through the resulting slums these visits were infrequent.

Adam looked out over the dimly lit city street. There was a dispersed crowd of people making their way towards the hospital, heading towards the pedestrian entry. The shine of a car’s headlights danced around the corner and dazzled Adam for a second as it caught him at the right angle. The Mk IV Vehicle Disabling System mounted behind concrete barriers to his left began to track the vehicle and feedback information to Adam’s heads up display on his ballistic goggles.

Driver: None

Registration: Expired 18 JUN 2035

Owner: Mr Albert Launceston (ERROR: Deceased)

Mass: Dry – 1044kg  

Suspected Load: 1100kg

Sending universal vehicle slow signal… Error: Rejected

Attempting Manufacturer override… Error: Rejected

Warning: Spectroscopic analysis indicated possible nitrate compounds, recommend swab for explosives

Passengers Detected: nil

Warning: Vehicle speed increasing

The sharp whine of electric motors was suddenly overpowered as the vehicles supplementary ethanol combustion engine kicked on. The vehicle lurched forward on its sagging suspension.

“Private Carlson, zap the tyres.” Adam ordered to the Mk IV’ operator. An orange confirmation light came into view on Adam’s HUD. He looked at is and blinked twice to confirm the Mk IV non-lethal activation. The Mk IV’s laser radiation warning light flashed yellow. A faint purple glow was visible as the Mk IV’s UV laser exposed the vehicles tyres to a century’s worth of sun exposure in milliseconds. The Mk IV quickly jerked from side to side as it targeted each tyre. The rubber duplicate flew off the tyres, but the vehicle only slowed slightly. A zoomed in view from the checkpoints cameras appeared on Adams HUD. The vehicle was now bounding along on concreted rims as it tore up the asphalt, its tyre cavities apparently having been filled with the stuff

Adam quickly cycled through the remaining disabling options he had left. He could switch the Mk IV’s radiation to microwaves and boil away its engines coolant and hope it seized, but there was the potential he’d ignite its ethanol fuel tank. He’d have to risk kinetic.

After a quick series of practiced menu authorisations the Mk IV’s warning light flashed red. A warning tone blurted before a sharp crack rang out as it loosed a single solid tungsten slug the size of Adams fist from its supplementary railgun. The rounds placement was calculated to have a 98% disablement probability. The MK IV had known the critical points of the vehicle from the moment its cameras selectively deciphered its image and its processors intelligently matched it thought its machine learning algorithm to its manufacturer’s data.

The round struck and destroyed the cars CPU. It did not slow. Almost instantly the MK IV recognised this and requested permission to fire at consecutive vulnerabilities at will. This decision from CPL Jacobson took what seemed an age to its optimised processing cycles during which the MK IV developed firing solutions to disable the vehicles main electrical power coupling and mechanical drive shaft. Actions that would disable to vehicle with minimal risk of fire or explosion. After several billion cycles the approval came.

Adam’s wetware was still processing the results of the Mk IV’s requests. Its first round had failed to disable the vehicle despite reporting its CPU destroyed. Its hypersonic spherical round designed to disable protected military vehicles had ripped a dinner plate sized hole in the cars fragile chassis and continued careening down the street out of sight.

The next two had had more local effects. One had lodged itself in a public hydrogen refuelling station and the other had bounced off the road and collided with the windows of a nearby high-rise sending a shower of glass onto some passer-byes below. They had taken cover where they could but Adam could see one woman staring at a large gash in her leg as the surrounding tissue started to trickle blood.

Then there was also the vehicle suspected to be laden with explosives now motionless in the street, having rolled to a stop 50 m away from the checkpoint, its engine screaming madly as whatever contingent processor that controlled it failed to comprehend the severed drive shaft. Multiple situations and only a handful of soldiers to deal with it. Adam then noticed a faint green light pulsing on his HUD indicating a message from command. He quickly opened it to view its contents.

In the hardened command centre deep underground in an undisclosed location LT Andrea Mickleham had been watching the events at CPL Jacobson’s checkpoint unfold. As she was doing this her assigned assistive AI was generating its own response plan to the situation. From her initial appointment Andreas training had been focused on decision making. She had no dedicated team, no set unit and no specific allegiance to any particular corps. But her training was focused towards tactical combat decisions, and her one job boiled down to reviewing information, making a decision and checking this against the “smart” AI that had been working alongside her since her appointment.

The goal of this relationship was to emulate the assigned human’s decision making cycle into a reproducible form. A form that didn’t have to sleep. A form that wasn’t affected by emotion, could never become angry or overwhelmed. Andrea’s AI had almost succeeded in replicating her decision making cycle perfectly. Very rarely was there alternate outcomes from her analysis to its.

Reviewing the data of CPL Jacobson’s situation, she did not know if this event was happening now, in the past or was entirely fabricated for the learning experience. The outcome would be the same: development of orders to deal with the situation. After a few seconds of review of her and her AI’s orders and, noting no major dissimilarities, she approved them for distribution. The only feedback she’d get was whether or not a successful outcome had been achieved at the end of her shift. In the meantime though another situation had already presented itself and her AI was already chugging away to a solution. Not wanting to give it anymore of a head start she dived in herself.

CPL Jacobson skimmed the orders. Little bite sized pieces that would help him deal with the reality he was facing now. An Explosive ordinance disposal team was on the way with deployable barriers to channel any resulting blast upwards. Priority now was for the clearance of civilians from the suspected blast zone which was his job as the current unit on the scene. Adam rallied his team. “Abrahams, Udesky, and Carlson maintain security here at the checkpoint. Everyone else, on me for casualty care.”

Adam and his detail moved over to the woman with the injured leg. His body cam sent her image to the national database and quickly discerned her identity. Details regarding her medical history we made available. She had no allergies or underlying conditions. The injury was large but not deep. A snap bandage was applied, automatically tightening around the wound to an appropriate pressure. More intensive wound care would be conducted once she was clear of the danger area.

Looking back towards the hospital, Adam could see the approaching red and blue lights of a tirage cart approaching the checkpoint. They assisted the woman to walk over to the autonomous cart. As they were helping her get in, 22 Engineering regiments deployable explosive containment barriers rounded the bend and positioned themselves around the vehicle. They began to pump their created seclusion zone full of their explosive neutralisation compound. The vicious liquid would slowly seep into every crevice of the vehicle and render any explosive safe. Its by-product was a lot of heat as the energy in the chemical bonds of the explosives was slowly released as they broke down. The vehicle would be nothing but a pile of slag by the end of it and the road would likely need resurfacing after clean-up but it was much safer than a controlled detonation.

Adam ticked off the tasks in his orders and they faded slowly from his view after turning green. Back to a quiet night on the checkpoint.

20
Cite Article
Harvard
APA
Footnote
RIS
(Hatcher, 2021)
Hatcher, S. 2021. 'Assistive AI'. Available at: https://theforge.defence.gov.au/article/assistive-ai (Accessed: 19 December 2024).
(Hatcher, 2021)
Hatcher, S. 2021. 'Assistive AI'. Available at: https://theforge.defence.gov.au/article/assistive-ai (Accessed: 19 December 2024).
Samuel Hatcher, "Assistive AI", The Forge, Published: November 15, 2021, https://theforge.defence.gov.au/article/assistive-ai. (accessed December 19, 2024).
Download a RIS file to use in your citation management tools.
Defence Technical Social

Comments

Disclaimer

The views expressed in this article are those of the author and do not necessarily reflect the position of the Department of Defence or the Australian Government.

This web site is presented by the Department of Defence for the purpose of disseminating information for the benefit of the public.

The Department of Defence reviews the information available on this web site and updates the information as required.

However, the Department of Defence does not guarantee, and accepts no legal liability whatsoever arising from or connected to, the accuracy, reliability, currency or completeness of any material contained on this web site or on any linked site.

The Department of Defence recommends that users exercise their own skill and care with respect to their use of this web site and that users carefully evaluate the accuracy, currency, completeness and relevance of the material on the web site for their purposes.