Self-driving cars are a hot topic right now, with the government recently passing a new bill called the 'Automated Vehicles Act' (AVA) which will see the 'safe' integration of partially and fully autonomous cars into our society over the coming years . Sadly, that does not include aerial mobility helmed by RoboCop Uber drivers.

But the AVA does mean a big change is on the horizon, and you'll be affected in one way or another by it. So what is it, and why should you care?

Let's wind back a bit. In 2018, the Center for Connected and Autonomous Vehicles (CCAV) requested permission from the lords and ladies in Big Ben to conduct a series of studies to build a case for how autonomous vehicles could improve road safety.

The reason? Perhaps their AI overlords asked them to, in what would be their first step towards world dominion. But the more likely cause is because previous research found that 88 per cent of road accidents are down to human error. As a result, those studies conducted by the CCAV have produced enough evidence to suggest that self-driving cars can bring this figure down.

How? Essentially, the new AVA acts as a set of guidelines that every manufacturer or OEM must follow if they want to sell a partial or full self-driving vehicle in Britain. To qualify, your product must go through a series of tests put forth by an independent body.

The crucial part of this bill is that it will retain the 'flexibility' required to accommodate any future developments with self-driving cars. And given that we're only going to see bigger and more ambitious developments, it means the AVA must evolve accordingly.

Before you start to dream of a future where your morning commutes include a bucket of custard creams and a car which automatically beeps at that moron who just cut you up, there's a bit more you need to understand about the wider subject. Let's start by simplifying the main one.

How is a self-driving car defined in the UK?

This remains the big gray area, but the Government's official answer is as follows: “One that has at least one self-driving feature, delivering sufficiently high levels of automation that it meets a legally defined threshold and is capable of safely driving itself with no human input. Such features could provide self-driving capability for all or part of a journey.”

Vague, isn't it? By that definition, a car with advanced cruise control, such as a Mazda 3, would be as much a self-driving car as a Tesla Model X, despite there being a clear difference in self-driving capability between the two. Therefore, any vehicles which pass through the AVA will be further divided into one of two sub-categories: 'user-in-charge' (UIC) and 'no-user-in-charge' (NUIC).

A UIC vehicle can be driven by a human or itself, for some or all of a given journey. Under the AVA, it means that the owner isn't held accountable for how the car behaves in its self-drive modes. However, in the instance that a UIC encounters a situation it cannot resolve itself, the driver will need to resume control and assume responsibility.

It's not yet clear what could cause this to happen, but we should find out more as the AVA develops. These cars will also be owned and maintained by their respective owners, who'll approach it in much the same way we do currently by ensuring it's insured, serviced and every cubbyhole is brimmed with sweets.

Alternatively, a car could be classed as a NUIC vehicle, which means every journey it completes would be done without *any* human intervention. It might complain if you drop a custard cream, though. These cars will be 'monitored' by a licensed operator, who will be in charge of how the vehicle runs, works and is maintained. Essentially, think of these as a sort of Hammersmith and City line that's compressed into the size of… a car.

For some peace of mind, if a car hasn't passed the AVA's guidelines and therefore been administered into one of the above two categories, it would be classed as a criminal offense for a vehicle to be sold as a self-driving car. Now onto the big one.

Who's responsible if something goes wrong?

This is where it gets a little tricky to explain, as the last section may have proven. Provided a car has passed the AVA and qualified for one of the two categories, and provided it is driving itselfall responsibility will be ushered away from the driver. That could include the manufacturer or a third-party organization such as a software developer. Though if the driver is in full control of the vehicle at the time of an incident, they may find themselves face-to-face with AI overlord 'V' at Southwark Crown Court.

But let's assume the driver isn't, and the car has made a mistake on its own. Here, the reason the blame is shifted away from the driver is that the car would've passed the AVA, which means its technology should have provided a level of safety that is at least comparable to that of a 'careful and competent' human. So if something does fail, none of the occupants can be blamed, and therefore the onus falls elsewhere.

And finally, when will the AVA bring self-driving cars to UK roads?

Given that the bill has only just been passed, it's too early to give a definitive answer here. But things are already in motion: UK-based start-up Wayve has just secured a billion dollars in funding to continue to develop its self-driving technology. It, alongside autonomous software brand Oxa, has already begun testing around London and Oxford. No mention of Coventry's treacherous ring road as yet, though.

So if everything goes according to plan, we could see the very first AVA-approved vehicles hit British roads as early as 2026. Exciting times ahead for overlord V and his consortium, then.