Understanding Artificial Intelligence
To find what we understand as Artificial Intelligence, we can go all the back to the 1960s. Allen Turin with computer science started to coin computer science and artificial intelligence terms and so we had these waves of what we call “AI Winters”, so we went from the 1960s which were the first AI winter, then 1980s, late 1990s, now we are in one of the latest waves. Therefore, it comes in cycles.
However, I think one of the key differences THIS time is that there is a convergence of not just a ton of data but power. We have the processing power to process at real time. However what critically was missing in the previous eras of artificial intelligence booms is that there are business cases now. The actual value is being generated and this value, people are willing to pay for it. I feel that was the missing piece previous winters/booms that is different this time. These days there are many cases where AI converges with other technologies such as AI and big data.
The Intersection between IOT, Big Data, and AI
One of Fresco Capital portfolio companies called Compology based out of the US does not use the term AI but if we dig deep, there are inverse correlations of how good a startups companies does AI and the number of times they use AI in their presentation material. Compology is at the intersection between IOT, Big Data, and AI. They build devices which have cameras, sensors, and GPS tracking specifically for waste trucks. The waste management industry has not been innovative for decades. What Compology did was that they allowed every single truck now to be connected so they know what their routes are, they know what they are picking up, and start gathering this data to optimize their routes.
In waste management, because the margins are so slim, a 2% efficiency increase is millions and millions of dollars. Now you install these devices and now you are getting a ton of data, through IoT and since it is all cloud-based, now you also have the processing power to actually process, analyze and start figuring out which route is the most efficient which creates so many more opportunities.
AI in Medicine
One of the problems that we face when it comes to medicine, however, is that AI currently creates a result. So it looks at patterns, let’s say, for example, detecting cancer in lungs. So the AI will look at millions of X-rays and starts to see patterns and then starts to fill out ideas on what are the correlations between certain structures of your lung and the likelihood of you getting cancer or having cancer. What it does it, it tells you a result. It doesn’t give you the “why” it thought of that. So as a doctor, the burden of fault is still on you. How do we overcome that? How do we adapt? These are some of the questions that will need to be answered when it comes to AI in medicine.
