Probably most of us believe that the onset of autonomous vehicles is inevitable, cars and trucks alike. But does anyone think it will go smoothly? I sure don’t.

For the most part we’re seeing very capable technology, and it will only get better.

But I fear that the mixing of autonomous trucks and other vehicles piloted by real people is fraught with risk. The reason is simple: people are stupid. The average motorist’s driving skills are nearly non-existent and there’s utterly no effort to improve them, and even in trucking, training is being improved only on a sadly limited scale. People are also unpredictable. They sneeze, they get distracted by billboards, they can’t resist texting — you name it, with their brains in neutral, they do it. So they often lose control, and I would challenge even the most sensor-laden autonomous truck to deal with all the resulting surprises.

So here’s an idea that comes out of a KPMG study that surveyed both car-maker executives and car-buying consumers as to the trends they see.

Of the executives, 74% believe that mixing autonomous and non-autonomous traffic (whether trucks or cars) will lead to severe safety issues. Mixed traffic is not possible, they conclude, saying it’s unlikely that fully autonomous vehicles and human drivers will use the same roads. As a result, they predict that we’ll see the trend of separation instead of integration, and with it the implementation of new road concepts and traffic systems over the next 10 years. Note: just 10 years!

That timeline doesn’t quite square with another conclusion about our future: 94% of those executives believe that a fully working and effective driving policy along with regulations for autonomous vehicles will be set up no later than 2040. I should think that, if we’re fooling with new road and traffic strategies that could start to be seen within a decade, we’ll want to have those other policies well in hand at more or less the same time.

Does any of that square with your own thinking? I’d love to hear what you think.

The KPMG report is the giant consulting firm’s 19th annual Global Automotive Executive Survey by its Automotive Institute. Bearing in mind that the survey focused on the automotive world, we’re not dealing with an exact mirror of what’s going on with commercial vehicles. Still, there are significant points of overlap.

Among other things, the report has confirmed my thinking about power sources of the near future.

I’ve written a few times in the last year or so that my money is on the hydrogen fuel cell as the eventual power source of the future, the one that will seize the biggest chunk of the internal combustion engine’s territory, and maybe not so long into the future. Nikola’s progress suggests that even long-haul heavy trucks will be able to go that route at some point soon. Clearly it will be plug-in battery-electric technology making the biggest progress in the near term, certainly for commercial vehicles in the lighter weight ranges and for a lot of medium-duty trucks. No disagreement on that one.

Executives in the KPMG survey project this kind of split by 2040: battery electric vehicles (26%), fuel cell electric vehicles (25%), internal combustion engines (25%), and hybrids (24%). That’s about cars, remember. I think you can almost discount hybrids in our trucking world.

Interestingly, KPMG respondents said: “Fuel-cell electric vehicles have replaced battery-electric vehicles as this year’s Number 1 key trend until 2025.”

And this next one is especially interesting: of the executives surveyed, 55% still believe that pure battery electric vehicles will fail due to the challenge of setting up the required infrastructure.

Thought-provoking, I’d say.

And with that I wish you all a Happy New Year. May all your miles be smooth.