It recognizes faces. He knows what videos you might like. He may be able to recommend the best treatment specifically for your case. Artificial intelligence and its subset of disciplines - such as machine learning, natural language processing, and computer vision - are increasingly integrated into our daily lives, whether we like it or not. What was once sci-fi has now become routine research and development in university laboratories around the world.
Similarly, startups working on artificial intelligence technologies have increased their capabilities - and with them stock prices have grown. More than 30 of these companies are now valued at $ 1 billion, according to research from CB Insights. Private companies with a valuation of a billion until recently were rare, such that they were even nicknamed "unicorns." However, there are now more than 325 of them, which in the aggregate estimate translates into almost a trillion dollars.
AI startups account for 10% of the total list, and are growing rapidly in number: over the past 4 years from zero to thirty-two. Last year, 17 AI startups hit the billion dollar threshold.
But what's behind all this money?
Big brother is watching you
Let's start with the bad news.
Facial recognition is probably one of the most common uses of AI today. Basically, it's a technology that's been around for years by a man named Woodrow Bledsoe who used a RAND tablet to semi-autonomously match faces from a database. This was in the 1960s.
Promotional video:
Most of us today are familiar with face recognition as a way to unlock smartphones. But the technology has gained prominence as a law enforcement oversight tool, especially in China.
It's no secret that facial recognition algorithms developed by several AI “unicorns” from China - SenseTime, CloudWalk and Face ++ (also known as Megvii) - are used to monitor the country's 1.3 billion people. The police there even wear AI glasses for this purpose.
Chinese startup Yitu Technologies, the fourth in a billion dollars, is also developing a security facial recognition platform and developing artificial intelligence systems for healthcare. For example, its intelligent four-dimensional imaging system CARE. AITM can identify various inflammations in real time, including possible early detection of cancer.
Artificial intelligence: doctor
As Peter Diamandis points out, AI is rapidly complementing and improving healthcare and life expectancy. In this regard, he points to another unicorn from China: iCarbonX, which plans to use machines to develop individual health plans for each person.
Several unicorns in healthcare hardware are OrCam Technologies and Butterfly. The first Israeli company developed the MyEye wearable device for visually impaired people, which attaches to glasses. The device can identify people and products, and read text by transmitting information using discrete audio.
Connecticut-based Butterfly Network has completely revamped the healthcare market with a portable ultrasound machine that works with a smartphone.
“Orcam and Butterfly are amazing examples of how machine learning can be integrated into solutions that deliver incremental improvements to even cutting-edge technology in ultra-competitive markets,” said Andrew Byrnes, chief investment officer at Comet Labs.
Artificial intelligence: driver
There are two AI unicorns in the Comet Labs portfolio: Megvii and Pony.ai.
The second is one of three billion-dollar startups developing artificial intelligence technologies in self-driving cars, and the other two are Momenta.ai and Zoox.
Founded in 2016 near San Francisco (and headquartered in China as well), Pony.ai debuted its PonyAlpha self-government system last year. This platform uses a variety of sensors (LiDAR, cameras and radar) to navigate the environment, but its “sensor fusion technologies” simplify the task by selecting the most reliable sensor data for any given driving scenario.
Zoox is another San Francisco startup founded a couple of years ago. In late 2018, it received the green light from the state of California to become the first autonomous vehicle company to carry passengers on a pilot program. Meanwhile, China's Momenta.ai is testing Level 4 autonomy for its self-driving system. Autonomous driving levels are rated from zero to five, with the fifth level equal to the person driving.
The buzz around autonomous driving is currently at its peak, and Byrnes believes regulatory checkpoints will keep most self-driving cars idling for the foreseeable future. The exception will be China, which takes a "systems" approach to autonomous transport.
“If autonomous transport solves more serious problems, for example, with traffic, and attracts government support, it will have the potential for rapid development. That's why we believe Pony.ai will be the winner in this area."
Artificial intelligence: process automation
An AI-powered technology that only fans of the iconic Office Space classic will appreciate is Robotic Process Automation (RPA).
RPA companies take over the day-to-day back office work, like filling out invoices or processing insurance claims, and outsource it to bots. Then the smart part comes into play, as bots can manipulate unstructured data, such as text in an email or even videos and pictures, to perform a growing variety of tasks.
Automation Anywhere and UiPath, both old companies founded in 2003 and 2005, do this. And since 2017 alone, they have raised $ 1 billion in aggregate funding.
Artificial Intelligence: Cybersecurity
Cybersecurity is another industry where AI is driving investment in startups. Big names like CrowdStrike, Darktrace, and Tanium use a variety of machine learning techniques to protect computers and other IT assets beyond simply updating their antivirus to the latest version.
Darktrace, for example, draws inspiration from the human immune system. Its algorithms can "learn" the unique design of each device and user on the network, identifying emerging problems before things get out of hand.
All three companies are used by large corporations and governments around the world.
What's next?
The most expensive startup in the world, China's Bytedance, is currently valued at $ 75 billion. What can she do? Create news and 15 second viral videos. However, companies like Comet Labs are not investing in such companies.
Byrnes sees real value in startups that take "data-driven approaches to problems unique to unique industries." For example, Chicago-based unicorn Uptake Technologies analyzes data from machines, from wind turbines to tractors, to predict problems before they affect machinery. PingThings, a small company, is conducting a similar analysis for the energy services sector.
Ilya Khel