technology trends

If you thought that the disruption we see in cars was unique to the automotive sector, then think again. Car manufacturers are merely reacting to a much bigger megatrend in technological development towards systems that can think for themselves. 

The application in the automotive world is pretty obvious: get cars to drive themselves, and it will open up a host of economic benefits, enriching our society tremendously. Once cars can navigate to destinations without human input, it will save time, reduce accidents, and end the need for people to take personal ownership of their motor vehicles. Ridesharing services will explode, prices will collapse, and people can look forward to a bright future in transport. 

The fundamental technology at the base of the autonomous vehicle revolution, though, is not unique. It fact, it’s what many commentators are calling a “general-purpose technology.” If you can get machines to “think” then there’s potentially no end of tasks that they can perform. Driving is just one of potentially millions of applications.

The Cognitive Revolution 

The concept of machine learning has been around since the middle of the twentieth century. The idea was that if you could provide a machine with enough data, it would be able to recognize objects, interpret sounds, and even hold conversations, using statistical techniques. 

The problem was that humanity didn’t have the technology to collect and store the information needed to train these systems, so the concept died. In the 1960s, collecting and storing gigabytes of data was practically impossible. 

Changes in technology, however, led to the re-emergence of the technology. It was clear that other approaches to artificial intelligence just weren’t working. “ Expert systems where programmers code all possible states of the world by hand wasn’t viable. These systems were fragile and couldn’t ever operate outside the realm of their programming. They weren’t much use.

Once sensors and storage fell in price, machine learning became a real possibility. At the start of the last decade, tech giants like Google and Baidu began to dabble in it, having some considerable success. Then over the following years, the potential of the technology exploded, with more and more applications coming to fruition. Today we’re in a situation where we can imagine training these systems to perform a host of cognitive tasks, including driving, writing, translating, and operating robots and cobots. 

Disruption Across Sectors

Take law firm document management, for instance. Legal professionals currently have to spend hours trawling through case papers, finding legal precedents that will help them fight their clients’ cases. Right now, the process is manual, but imagine how you could apply machine learning. A document-reading algorithm imbued with intelligence could instantly scan millions of pages and then create a summarised document of its findings that a lawyer could then use in court. 

The problem was that humanity didn’t have the technology to collect and store the information needed to train these systems, so the concept died. In the 1960s, collecting and storing gigabytes of data was practically impossible. 

Changes in technology, however, led to the re-emergence of the technology. It was clear that other approaches to artificial intelligence weren’t working. “Expert systems” where programmers code all possible states of the world by hand wasn’t viable. These systems were fragile and couldn’t ever operate outside the realm of their programming. They weren’t much use.

Once sensors and storage fell in price, machine learning became a real possibility. At the start of the last decade, tech giants like Google and Baidu began to dabble in it, having some considerable success. Then over the following years, the potential of the technology exploded, with more and more applications coming to fruition. Today we’re in a situation where we can imagine training these systems to perform a host of cognitive tasks, including driving, writing, translating, and operating robots and cobots. 

Disruption Across Sectors

Take law firm document management, for instance. Legal professionals currently have to spend hours trawling through case papers, finding legal precedents that will help them fight their clients’ cases. Right now, the process is manual, but imagine how you could apply machine learning. A document-reading algorithm imbued with intelligence could instantly scan millions of pages and then create a summarised document of its findings that a lawyer could then use in court. 

Don’t think for a second that the automotive industry is the only place that machine learning technologies will disrupt. Driverless cars are low-hanging fruit and somewhere that the technology will have a significant impact, but this isn’t an isolated phenomenon. Artificial intelligence will probably be the story of the century.