AI-Series (0): What You Always Wanted to Know About AI and More
Do you know Paro? Paro is a robot that looks like a cuddly baby seal. It responds to sensory stimuli such as petting, and can recognize and distinguish between human voices. For several years, the robot has been used in therapy for dementia patients. The social robot was developed at the National Institute of Advanced Science and Technology (AIST) in Japan. What empowers Paro is artificial intelligence.
Smart algorithms have long been part of our everyday lives in many different areas. They search for relevant offers for the respective user in online stores, they control traffic lights and, in the future, autonomous vehicles, they support doctors in diagnostics, and they provide quantum leaps in preventive medicine. They are able to do so because they recognize signs of potential disease risks long before they become obvious. They control production within large industrial plants while taking into account market data such as demand and raw material prices in real time.
Many companies – industrial companies, financial service providers, e-commerce companies and retail companies – have been looking at artificial intelligence for several years. They are all looking for ways of using it to make processes more efficient, leverage value or develop new business models. The big challenge here is to transform a promising proof of concept into a live operation.
BCG Platinion has extensive experience with AI-based solutions. We want to share this knowledge with you. Over the next few months, we’ll be publishing a series of deep-dive articles around the topic of artificial intelligence and how businesses can benefit from it.
In the first article, we introduce you to the basics of machine learning – i.e., artificial intelligence – and provide an overview of the three common methods and their areas of application. Article two dives deep into core criteria and methods for the quality testing of AI models. In article three we present a method to succeed in minimizing bias to get realistic results from AI models.
