Now that Robotic Process Automation (RPA) is combined with AI bots in a no-coding environment where platforms provide turnkey solutions, SME’s will become change leaders, especially as cost barriers vanish. Let’s get up to speed with where it comes from and where it is heading next:
As recently as 2010, artificial intelligence (AI) was used in science-fiction movies only and no one ever suspected that it would be part of day-to-day life. Today research shows that even accountancy practices use it to drive audits.
Since 2010, AI gradually found its way into people’s lives, analyzing personalities in job interviews, helping people to take better smartphone photos and enabling people to make purchases without paying the cashier. It is now being used for beneficial ways, not-so-beneficial ways (e.g. deepfake videos) and controversial things (e.g. surveillance and facial recognition).
With technological advancements, combined with cheaper and easier access to powerful computers, AI has invaded many different parts of life in the last decade. Most of the AI that people mostly encounter relies on machine learning i.e. the ability of a computer to teach itself by intensely studying data. In the past decade, technological developments have been specifically focused on deep learning which mimics the way the brain works. Google figured this out in 2012.
This is just the beginning
AI is still unsophisticated. For example, a machine learning algorithm performs one function and requires loads of data to learn how to do it well. Scientists are still looking for a way for machine learning systems to be better at generalizing and learning from fewer examples.
As Pedro Domingos, a University of Washington computer science professor said, “Ten years ago, deep learning was not on anybody’s radar, and now it’s in everything… We’ve come a thousand miles, but there’s a million miles still to go.”
6 of the many ways AI has impacted lives
AI has impacted lives in many ways in the last 10 years of its journey. Here are six of the many ways it has affected lives.
AI is in most smartphones, e.g. facial recognition for unlocking the phone. Google and Apple are trying to run AI directly on handsets (with chips) so that activities such as speech recognition and translation of words can be performed on the phone instead of a remote computer in order to preserve data privacy. One example is Google’s new transcription app called Recorder which can record and transcribe in real time, can identify various sounds such as applause and music and can run on Google Pixel smartphones.
Facebook used to focus on connecting people when it started in 2004, but now they are obsessed with doing it with AI. After years of research and development, deep learning now supports everything including ads, posts and the way friends can be tagged in photos automatically. AI even knows how to remove hate speech from social networks but still has a long way to go especially in spotting hate speech and violence. Twitter, Instagram and other social networks also rely heavily on AI.
People interact with AI often when they talk to Apple’s Siri, Amazon’s Alexa and Google’s Assistant. These helpers notably understand what people are saying and respond with what they want. Virtual assistants began in 2011 when Apple deployed Siri on the iPhone and Google followed with Google Now in 2012, both limited to smartphones. Then Amazon’s Alexa, which came in 2014, helped the virtual assistant market to explode and brought AI to more homes. Google upgraded to Google Assistant in 2016. In the 3rd quarter of 2019, Amazon shipped 10.4 million Alexa-run smart speakers!
AI has improved much. Its capabilities as a surveillance tool have improved too. Facial recognition is its most controversial capability. It identifies people from live or recorded video and still photos by comparing features with those in a database of faces. It’s been used by police, at concerts, airports and other settings. Controversy comes from concerns about accuracy and privacy. In December 2019, a US government study found widespread bias in about 200 facial recognition algorithms and that racial minorities have high risk of being misidentified than whites. San Francisco, Somerville in Massachusetts and Oakland in California and other cities banned the use of the technology by city departments in 2019.
AI is gradually being used to diagnose and manage many kinds of illnesses e.g. identifying lung cancer, gastrointestinal issues and mental health issues. The work is still in early stages, e.g. Mindstrong Health uses an app to measure moods in patients struggling with mental health issues. Auggi is building an app to track digestive health problems for people with chronic gut-related problems. Seed Health is also building its own. Soon there will be many apps helping medical professionals and sick people.
The most amazing use of AI is in the arts though no one can say for sure whether a machine can have creativity. In the past 10 years, AI has been used to make paintings, musical compositions and more. The art can even make money as was seen in the 2018 auctioning of “Edmond de Belamy,” a blurry Old Masters-esque piece of art produced by a machine. The print was created using GANS, a cutting-edge technique.
Suggested reading by Avi Benezra: How RPA affects the travel and tourism industry.