From Horsepower to Brainpower: Who Wins When Cars Get Smarter?

November 18, 2026 | Michael McQueen

Not long ago, the idea of a fully driverless taxi navigating Australian traffic sounded like one of those predictions that was always five or ten years away. Then Alphabet’s Waymo began talking with Transport for NSW about testing its robotaxis on Sydney roads. The company has not announced an Australian launch date, and any deployment would still need regulatory approval. But the conversation itself is revealing. A technology that spent years living in prototypes and carefully staged demonstrations is edging towards ordinary streets.

At the same time, Tesla has begun producing a Cybercab designed without a steering wheel or pedals. Chinese vehicle brands are rapidly gaining share in Australia. Cars are increasingly updated through software rather than a mechanic’s toolbox. Insurers are beginning to ask whether autonomous systems should attract lower premiums than human drivers. And the dealership, once the unavoidable gateway to buying a car, is being challenged by digital-first models that let customers complete much of the transaction from a phone.

It is tempting to describe all of this as the EV transition. I think that misses the bigger story.

For more than a century, the automotive industry competed largely on mechanical capability: horsepower, cylinders, acceleration, reliability, fuel economy and engineering refinement. Those things still matter. But value is increasingly migrating from the mechanics of the vehicle to the intelligence inside and around it: software, sensors, AI, connectivity, autonomous systems and the digital ecosystem that follows the customer long after the car leaves the showroom.

We are moving from horsepower to brainpower.

The electric motor is only part of the story

Electrification matters enormously, but one reason it can be misleading as the headline for automotive disruption is that replacing an internal-combustion engine with an electric drivetrain does not, by itself, explain why the competitive order is changing so quickly.

The more consequential change is that a modern vehicle is becoming a software platform on wheels. McKinsey’s 2026 mobility research found that consumers are placing increasing emphasis on sophisticated vehicle technology and advanced driver-assistance systems, while traditional differentiators such as comfort and quality are becoming less decisive. That creates an opening for companies that do not carry a century of automotive heritage but are very good at batteries, software, interfaces and rapid product iteration.

Australia offers a useful glimpse of this shift. In the first half of 2026, Federal Chamber of Automotive Industries data reported by ABC News showed BYD‘s share of the market more than doubling from 3.8 per cent to 8.6 per cent among brands captured by the dataset. The change is not simply about Australians deciding they want an electric car. It is also about a wave of newer manufacturers entering the market with different product cycles, technology stacks and price points.

That disruption has a less glamorous side. The same ABC investigation documented dealer concerns about delays securing spare parts and EV batteries from some newer entrants, along with a mismatch between the legal responsibility dealers carry under Australian Consumer Law and the practical control manufacturers often retain over defects and warranty decisions. The Electric Vehicle Council disputed any suggestion that the problem applied broadly across manufacturers, arguing that members were confident in their parts supply.

That disagreement is worth noting because it points to a larger truth. The future of automotive will not be won on the specification sheet alone. A brilliant vehicle with a frustrating repair, warranty or software-support experience is still a frustrating vehicle. As cars become more technologically sophisticated, the ecosystem around the car becomes part of the product.

The car is learning to drive itself

Autonomy is where the move from horsepower to brainpower becomes most visible.

Waymo already operates fully driverless commercial services in several US cities and has announced plans for London. Tesla is pursuing a different technical and commercial path, with its purpose-built Cybercab. McKinsey’s latest industry work suggests robotaxis are likely to be the first major Level 4 autonomous-driving application to scale, ahead of privately owned cars that can drive themselves almost anywhere.

This is important because we have spent years imagining autonomy through the wrong lens. The popular image is that one day everyone’s existing car suddenly becomes fully self-driving. The more plausible near-term transition is narrower: autonomous systems become very good inside defined operating environments, then expand city by city and use case by use case.

Even that timeline has proved stubborn. McKinsey’s survey of autonomous-driving industry leaders found expected adoption timelines had slipped by one to two years across many use cases compared with its 2023 survey. Large-scale robotaxi deployment was pushed to around 2030 in the survey, while Level 4 urban pilots for privately owned passenger cars moved further out.

That is why I would be wary of anyone offering a precise date for the arrival of the driverless future. The direction is becoming clearer. The timetable is not.

The safety question may soon reverse

The most consequential development may not be that autonomous cars can drive. It is the growing evidence that, in some environments, they can already drive more safely than we do.

In July 2026, the Insurance Institute for Highway Safety compared Waymo robotaxis with human-driven vehicles across San Francisco, Phoenix, Los Angeles and Austin. The analysis found Waymo vehicles were involved in 68 per cent fewer police-reportable crashes per mile. Single-vehicle crashes were 85 per cent lower and injury crashes 81 per cent lower on a per-mile basis.

Those numbers deserve caveats. Autonomous-vehicle crash reporting remains inconsistent, and researchers lack comparable mileage data for most operators. IIHS had to exclude a substantial portion of reported incidents because of duplicates and other data-quality issues. Austin also produced a different result, although the sample there was small. This is encouraging evidence, not a final verdict.

There is another complication. In early 2026, a Waymo vehicle struck a child who ran into the road from behind a parked SUV near a Santa Monica school. Waymo said the vehicle braked immediately and reduced its speed dramatically before impact, and police reported no injury. The company’s modelling suggested an attentive human would have struck the child at a higher speed. Safety experts nevertheless argued that reaction time was only part of the question: a skilled human might have anticipated the school-zone risk earlier and adjusted their behaviour before the emergency arose.

That is exactly the sort of uncomfortable ambiguity we should expect during this transition. Machines can outperform humans on some dimensions of driving while still behaving in ways we find strange or inadequate in others.

But the debate may gradually invert. We have spent years asking, ‘Can a machine be trusted to drive as safely as a human?’ If independent safety evidence keeps moving in the same direction, the question could become: ‘When does choosing the human driver become the riskier option?’

Trust will lag capability

Even if the statistics eventually become compelling, public acceptance will not automatically follow. Australian research cited in reporting on Waymo’s local ambitions found 47 per cent of surveyed drivers regarded autonomous vehicles as a desirable travel option, yet 74 per cent still wanted the ability for a human to take control. More than four in five expressed uncertainty about who would be legally responsible when something went wrong.

That gap between capability and comfort matters. We do not evaluate machines the same way we evaluate people. A human driver can make a catastrophic mistake and we tend to treat it as an individual failure. When an autonomous car makes one, we are more inclined to question the entire technology.

There are good reasons for a high standard. Autonomous systems are deployed at scale, which means a software flaw can be replicated across a fleet. But demanding perfection from machines while accepting tens of thousands of human-caused road deaths each year creates its own inconsistency.

Trust will probably be built less through advertising than exposure. The first autonomous ride is psychologically significant. The tenth is likely to feel far less remarkable. Technologies often become normal not when people understand every technical detail, but when repeated experience makes the unfamiliar boring.

The dealership is becoming a technology business too

The redistribution of value is not confined to the vehicle. It is changing how cars are sold. Carvana built its US used-car business around an online buying experience, home delivery and the memorable glass car ‘vending machines’ that became its physical calling card. In 2026, its move into new-car dealerships began unsettling traditional dealers.

The interesting part is not whether Carvana ultimately dominates new-car retail. It is what its success has already demonstrated: plenty of consumers will spend tens of thousands of dollars on a vehicle without following the traditional dealership ritual of repeated visits, commissioned sales conversations and price haggling.

This does not mean physical dealerships disappear. Cars remain expensive, tactile products that need servicing, repairs, trade-ins and local support. In fact, the problems some Australian consumers have encountered with parts and warranties are a reminder that a physical service network can become more valuable when the technology becomes more complex.

But the role changes. A dealership built primarily around information asymmetry and transaction friction looks increasingly vulnerable when the customer arrives with transparent pricing, finance comparisons, vehicle reviews and an AI assistant in their pocket. The opportunity is to become the trusted physical layer of a much more digital relationship.

Insurance follows the intelligence

There is a similar shift coming for insurance. Traditional motor insurance is largely built around human risk: age, driving history, location, kilometres travelled and the statistical likelihood that a person will have an accident. As more of the driving task moves into software, risk starts migrating with it.

A genuinely autonomous vehicle raises different questions. Was the crash caused by the passenger, the vehicle manufacturer, a sensor failure, a software update, a mapping problem, the fleet operator or a cyberattack? The answer matters because the economic centre of motor insurance could gradually move away from individual driver behaviour and towards product liability, software performance, fleet risk and cybersecurity.

This will not happen overnight because human-driven and increasingly automated vehicles will share roads for years. But it is another example of the same underlying shift: once intelligence moves into the machine, industries built around human behaviour have to rethink where risk and responsibility sit.

The future arrives at different speeds

There is a tendency in technology forecasting to imagine that once something becomes technically possible, adoption follows in a smooth upward curve. Automotive is a good antidote to that thinking.

A vehicle has a long replacement cycle. Roads and regulation move slowly. Charging infrastructure is uneven. Insurance frameworks need updating. People form emotional attachments to cars and to driving. Rural mobility looks different from dense urban mobility. A robotaxi that works brilliantly inside a mapped section of San Francisco does not automatically solve a trip on a regional Australian road.

Even the transition technologies are messier than their labels suggest. Research in the latest European data on plug-in hybrids found real-world emissions benefits substantially below the assumptions used in laboratory and regulatory calculations, largely because drivers did not operate the vehicles electrically as often as models assumed. A technology can be capable of producing one outcome while real human behaviour produces another.

This is why the automotive future is arriving both faster and slower than it appears. The breakthroughs are fast. System replacement is slow.

Waymo can operate a commercial driverless service while most Australians still drive conventional cars. A Cybercab can roll off a production line without a steering wheel while regulations written around steering wheels remain a barrier. Chinese brands can gain share quickly while building the service networks needed to support vehicles over a decade takes much longer.

From horsepower to brainpower

For the past century, we have largely understood automotive progress through the machine itself. More power. Better engines. Greater reliability. Improved efficiency. The next era asks us to widen the frame.

The car is becoming a node in a much larger intelligent system. It senses its environment, learns from data, receives software updates, communicates with infrastructure, assists the driver and, increasingly, makes decisions on the driver’s behalf. Around it sit digital sales platforms, charging networks, insurers, mobility services, repair ecosystems and regulators trying to keep pace.

That does not make mechanical engineering unimportant. A car still has to stop, steer, survive heat and rain, protect its occupants and last for years. But the centre of competitive gravity is shifting.

For automotive leaders, the question is therefore bigger than which powertrain wins. It is where value migrates when intelligence becomes one of the vehicle’s defining features. For dealers, it is what customers still need from a physical network when the buying journey becomes digital. For insurers, it is who carries risk when software does the driving. For governments, it is how to regulate systems whose capabilities change through updates rather than model-year redesigns.

And for the rest of us, it may eventually change what we mean by driving at all.

The automobile transformed the twentieth century because it gave people unprecedented freedom of movement. The next transformation may come from separating mobility from the need to personally operate the machine that provides it.

The engine was the defining technology of the old automotive age. Increasingly, the intelligence around it will define the next one.


Michael McQueen is a globally recognised trend forecaster, change strategist and keynote speaker.

A bestselling author of 10 books, Michael’s latest release was named by Malcolm Gladwell and Adam Grant as one of the top five must-read new leadership books. He is a sought-after media commentator, with his insights regularly featured in Forbes, The Guardian, and CNN.

To find out more about Michael and his work, click here.

NOTES

Insurance Institute for Highway Safety study: https://www.iihs.org/news/detail/waymos-driverless-cars-crash-less-often-than-people

McKinsey Center for Future Mobility. “How Consumers Are Reshaping the Future of Mobility.” McKinsey & Company, July 2026.

Kelkar, Ani, Kersten Heineke, Martin Kellner, Charlotte Wiemuth, and Eduardo Mañas-Pont. “The Future of the Autonomous Vehicles Industry.” McKinsey & Company, January 2026.

Swan, David. “Google Wants to Bring Its Driverless Taxis to Sydney’s Streets.” The Sydney Morning Herald, November 14, 2025.

Libatique, Roxanne. “Waymo in Talks to Test Driverless Taxis on Sydney Roads.” Insurance Business Australia, November 17, 2025.

Wilmot, Stephen. “Driverless Taxis to Take On London’s Storied Black Cabs.” The Wall Street Journal, October 15, 2025.

Peterson, Becky. “Tesla’s Grand Plan for the Future Is a Car With No Steering Wheel.” The Wall Street Journal, March 12, 2026.

ABC News. “Dealers Warn EV Drivers of Repair and Refund Delays under Warranty from New Brands.” July 17, 2026. Market-share figures sourced by ABC from the Federal Chamber of Automotive Industries; Tesla and Polestar were not included in that FCAI dataset.

The Wall Street Journal. “Carvana Grew Into a Used-Car Titan. Its New-Car Sales Project Has Dealers Rattled.” May 19, 2026.

International Council on Clean Transportation and related European real-world plug-in-hybrid research: https://theicct.org/pr-plug-in-hybrid-emissions-gap-widens-despite-longer-electric-ranges-icct-study-shows/

Continue reading this series

From Horsepower to Brainpower: Who Wins When Cars Get Smarter?

Gen Z Has Grown Up. Our Assumptions About Them Need To As Well.

When AI Meets HR: Welcome to the New World of People Leadership

The Lawyer’s New Value: From Legal Answers to Better Judgement

AI Won’t End Accounting but it Will Change What Accountants Are Worth

Beyond the Chair: Seven Forces Reshaping Hair and Beauty

Ready to get started?

If you like what you’ve heard and are ready to take the next step, we’d love to hear from you!

START NOW

Change your team dynamic with this [GUIDE]

DOWNLOAD

Related Posts

Radical Curiosity by Michael McQueen, book cover
Launching February 2027

Michael’s new book Radical Curiosity

Unlock the Power of an Open Mind

Don't miss your chance to pre-order Michael's new book Radical Curiosity: Unlock the Power of an Open Mind

SPECIAL BONUS: Pre-order today and get the 10-day Curiosity Kickstart video masterclass series FREE (RRP $39.95).

Download our [name of download]

Ex ipsum culpa mollit enim enim. Excepteur duis esse in esse aliqua eiusmod.

Download Form