There’s a lot of talk about IoT and AI, so in this post we will cover the basics – what is AI (Artificial Intelligence)?
One of the definitions says: Artificial intelligence (AI) is an area of computer science that emphasizes the creation of intelligent machines that work and react like humans. AI is academic discipline since 1956. and it had its ups and downs – today it’s gaining popularity as great improvements were made in computing power, storage, big data and advanced algorithms. Traditional AI research are focused on: reasoning, problem solving, knowledge representation, planning, learning, natural language processing, perception, ability to move objects. Since AI is a rather complex field, we have diversity of experts involved in its development: computer scientists, information engineers, mathematicians, psychologists, linguists and many others – all focused on mapping the human intelligence in the way that machine would be able to mimic. The holy grail of AI is data, it will give you win in a competitive industry – everybody will be applying similar techniques and the best data will give the right answers. How to get the data? Here we have IoT as a right tool for gathering the data in large volumes and feeding to the data hungry AI.
Where can we apply AI? Well, wherever we have an intellectual task, AI is up for it, so we have a broad set of use cases, I’ll just name a few:
- Healthcare – for example a doctor’s assistant that will memorize all the relevant data and suggest the most effective patient treatment
- Autonomous vehicles – for example self driving trucks are in testing phase, but soon this is something we’ll see quite often on the roads, especially on highways
- Economy and finance – prevention of fraud and financial crime is an example of today’s AI usage and something that will definitely advance in future
- Military – even though this is some kind of dark side of AI usage, there are many examples of it. Like drones that we can see in action already today. But who can say for sure that AI soldiers ready for the battle already don’t exists? Just remember that they had developed intelligent personal assistant in 2003. – long before Alexa, Siri or Cortana were developed.
- Advertising – AI platforms that will predict behavior of a person and offer personalized promotions, recommendations and purchase option. Are some of those things already seem familiar to you?
- Manufacturing – AI will be used to forecast load and demand as well as need for service and maintenance, all based on IoT data, from the factory itself as well as from supply chain and consumer.
Where do we see AI usage today? To help you get a feeling what is AI and what can it do today, I’ll mention some of the most widely spread examples:
- Apple’s Siri personal assistant – it uses machine learning technology to learn new things and understand our questions about information we need, direction where we want to go, it can log events in our calendar, etc.
- Amazon’s Alexa smart home assistant – the original idea was to built a smart home hub, but its ability to unscramble speech even when you are far away from it in a room, made it popular for getting the needed information on the web, setting the alarm, setting calendar events, making a shopping list, etc.
- Tesla – unfortunately I don’t own one and I also didn’t have a chance to try it, even though I can often hear that it’s a must car for everyone that’s into technology and cars. Tesla cars have predictive capabilities, self driving features and they are getting smarter and smarter by every new update done automatically over the air.
- Netflix – it uses predictive technology to suggest movies to us based upon our previous choices and reactions by analyzing a huge database of records.
Even though general idea of AI is to help us in so many ways, there’s a lot of worries around it: will we be able to control its progress, will people lose their jobs because of it. There are many world leaders whose statements are also worrying – “Whoever becomes the leader in AI will become the ruler of the world” statement by Vladimir Putin. Elon Musk often emphasizes how the development of AI is dangerous if it goes out of our control and we have similar concerns stated from Bill Gates, Steve Wozniak and Stephen Hawking.
Well, AI is not that perfect, today it still lacks “common knowledge” that humans have, so mistakes that AI often does for us humans will seem incomprehensible. AI systems of today are not multidisciplinary as us people are, they are specialized per single tasks (internet search, facial recognition, driving a car).
Personally I see AI as a partner assistant helping us to do better by giving us clearer vision and understanding, greater memory, etc. It will not replace us, but during the development we definitely must be fully aware what it can do if it goes out of control and thus we need to align the goals of the AI with ours before it outsmart us. If you’d like to learn more about AI, you can chose some of the online courses: like Artificial Intelligence (AI), Intro to Artificial Intelligence or many others that you can find online.