Investing in the Future of Intelligent Automation
Over the next 20 years, our lives will be radically transformed by intelligent automation. Much like how electricity fueled innovation and capitalism, current advances in artificial intelligence are going to give rise to a new era of profound technological and economic change.
Like humans, to gain intelligence, machines need to learn. The faster they can learn, the more autonomous they will become. They learn by consuming data and training themselves on this data. However, machines can learn MUCH faster than humans. As pointed out by blogger, Tim Urban:
“The brain’s neurons max out at around 200 Hz, while today’s microprocessors (which are much slower than they will be when we reach AGI) run at 2GHz, or 10 million times faster than our neurons. And the brain’s internal communications, which can move at about 120 m/s, are horribly outmatched by a computer’s ability to communicate optically at the speed of light.”
Estimates have gone as far as to predict that machine learning technology will contribute as much as “$15.7 trillion to the world economy by 2030” due to the impact of increased productivity and increased consumption due to better recommendations. Truly, every organization in the world stands to benefit from the power of machine learning.
How BRV Thinks About This Change
During Fund IV (2007), the smartphone started to proliferate and has since become universal. These phones are packed with incredible computing power and a vast array of sensors that can stream real-time data back to the cloud. Waze, from Fund IV, is a great example of a service that collected data from the phone’s sensors and relayed that data back to the cloud to help its users make more informed decisions.
During Fund V (2014), IoT sensors started to proliferate and are now becoming ubiquitous. These sensors are also connected to the cloud. Our Fund V investment in HumanAPI is a great example of a company that collects vast quantities of data exhaust that comes from these IoT wearables and uses that data to help its customers makes more informed decisions.
As for our current fund — Fund VI — we are now at a point where all the hardware is in place to contribute data back to the machines, for them to start learning and becoming more intelligent.
At the heart of intelligent automation is machine learning, which is enabled by three driving forces:
- Algorithms. While unquestionably important, algorithms have been around for 60+ years and have become, largely, commoditized. There’s limited value in the code.
- Compute power. Incredibly fast and affordable GPUs have paved the way for easy access to machine learning in the cloud. All the major cloud providers (Amazon, Microsoft, Google) have democratized access to these machine learning algorithms for all the world to use.
- Data. The most valuable aspect of machine learning. Without the data, there would be nothing to teach the machines. Data comes from everywhere. This includes pre-existing company data sets, as well as the data exhaust that comes from today’s mobile devices. Not to mention, there are all the inexpensive IoT sensors that are rapidly proliferating around the globe and contributing to the data. Everything from GPS, to motion sensing, to low-cost HD cameras that we can now place everywhere.
We want to capitalize on all the data coming from these devices, by investing in companies that are automating processes based on this data.
Some Changes We Expect to see by Industry
Transportation
Using advanced predictive models, both cars and trucks will be able to safely navigate roadways. This is one area where very fast and reliable cloud services will play a crucial role in vehicle-to-vehicle communication.
Healthcare
Thanks to all the wearables that have been contributing data to these machine learning algorithms, we are on the verge of powerful predictive healthcare systems. It’s only now that we have technology able to make an intelligent diagnosis, often, more accurately than human physicians. We’ve seen healthcare companies that will scan your retina, with off-the-shelf HD camera components. Those retina scans will then get sent to the cloud where they can be processed by machine learning algorithms that have been trained to identify early onset diabetes in patients. It’s cheaper and more accurate than physician diagnosis. This is only just now possible, due to the vast amount of training data (in this case, other retinal images) that has been collected from across the globe.
Insurance
The insurance sector, which is estimated to reach nearly $4.1t in underwritings by 2020, will leverage Intelligent Automation for the underwriting of long-term care and life insurance policies.
Financial Services
Predictive financial broker tools for both enterprise and consumer will help inform and guide investment decisions based on a plethora of real-time data, such a weather, international conflict, and a myriad of other not-so-evident interconnected relationships.
Regarding fraud detection, these systems, when given dates, times, merchant info and customer info, will be able to predict fraud with unprecedented accuracy.
Back Office
What organizations are quickly realizing is that nearly every single back office task will quickly approach full automation.
Since most back-office tasks are cost centers, it makes sense that this work is outsourced to third party machines. Once trained a machine can operate more efficiently than traditional human labor.
This will apply to everything from claims processing, to loan applications, to routine tasks such as scheduling. Nearly all repetitive, trainable back office tasks will be outsourced to intelligent machines.
Industrial
Expect fully autonomous factories with close to zero human presence. We should also see end-to-end order fulfillment without little to no human intervention with the help of robots that can grab items off the shelf to drones that can solve the issues of last mile delivery.
____
As intelligent automation proliferates through the U.S. and global economy, it will continue to drive economic growth and development, as all great technologies do.
