Automation has been in factories for a long time, but the recent emergence of learning machines – so-called artificial intelligence (AI) – has opened up a whole new world for today’s companies.
Machine learning is a software capability that enables a program to keep improving its performance without a human operator. In just the past couple of years, machine learning has become significantly more effective and accessible.
While AI can do a lot for business, it has raised a lot of unrealistic expectations. It’s important for companies looking to invest in automation to realize what current technology can do – and what it can’t.
What is today’s software capable of?
The greatest developments have been in two main areas: perception and cognition. Perception advances have been particularly significant in relation to speech. Voice recognition software still isn’t perfect, but countless people are now using it with digital assistants like Siri, Alexa and Cortana.
A recent study from Stanford University said speech recognition is now around three times as quick as typing on a cell phone. Amazingly, the error rate has dropped from 8.5 percent to 4.9 percent in the span of just one year; from 2016 to today.
Image recognition has also improved drastically. You might have seen that Facebook can now identify your friends’ faces in the photos you post. In the workplace, image recognition is currently replacing ID cards.
The second big advance has been in cognition. Google’s DeepMind team has used AI to enhance the cooling efficiency at data centers by greater than 15 percent, even after they had already been optimized by human experts. Other uses of advanced cognition capabilities include cyber security businesses using it to recognize malware, and PayPal using AI to stop money laundering.
Many AI systems are now superior to humans at many tasks. This development has opened up substantial possibilities for changing the way business is done. Once AI-based systems exceed human efficiency at a given task, the system is more likely to spread. For example, drone makers have leveraged enhanced vision systems to automate some responsibilities of security guards.
AI and the average business
AI isn’t just something that’s relevant to tech giants like Google or deep-pocketed multi-billion-dollar corporations. The technology is currently at a point in its evolution that most average businesses can take advantage of it.
First of all, AI skills are being formally taught in classrooms across the country, and that means people are coming out of these classes with useful skills. Second, AI software and hardware is available commercially, for rent or purchase.
Third, AI can have profound positive effects on a business without mountains of data waiting to be analyzed. These systems learn as they go, so a system can significantly improve performance as new data is recorded and analyzed.
Also, AI systems can quickly spot patterns in a modest amount of data that would take humans far too long to pick out. For example, a company could use a machine-learning solution to scan customer service chats to see which responses generate the best customer satisfaction responses.
At SSi People , we provide our client companies with access to professionals with the latest and greatest tech skills. Please contact us today to find out how we can help your company.