10 years down the line it will be a different story though
I know you were talking about cars, but I think that's the important line.
The vast majority of aircraft are and will remain manned
That's factually incorrect. Googling around suggests that there are about 20,000 airliners in operation globally, perhaps 300,000 GA aircraft, and 100,000 military aircraft. In contrast they sell well over a million serious-hobbyist drones a year, and large numbers of radio-controlled aircraft and vast numbers of smaller flying toys. The USAF now has more drones than manned aircraft as do the British military. There are apparently 1600 remote-control crop-spraying helicopters in Japan - the RMAX which I believe is the most common one has an AUW of about 90kg and costs about $90,000 so they're not toys.
The number of flying hours is a different question, but when I was seriously into r/c helicopters my annual r/c flying hours were rather higher than I have averaged with my PPL so I doubt it's trivial.
Drones are already here in large numbers: what's in question is how long it's likely to take for autonomous drones to become widely adopted.
People's lives being priceless ahead of profit or convenience of UAV operation, legislation must reflect this.
Lives are important, but consider that about 60 light-goods-vehicle drivers are killed every year in the UK and 600 seriously injured. They also kill about 60 cyclists. I can't be bothered to find the stats for pedestrians and car-drivers, but suffice to say that there's a lot of truth in the stereotype of the 'White Van Man'.
The most dangerous roads are fast rural roads, so in remote areas such as the Scottish highlands and islands where the roads are dangerous and people sparse, you could have quite a high drone accident rate and still harm fewer people than if you sent the goods by road.