Human-Centered AI: Artificial Intelligence That Works With Humans, Not Against Them
Author | Min Sun, Chief AI Scientist, Appier
The way we think about artificial intelligence (AI) is changing. Traditionally, AI has tried to emulate and surpass human intelligence by mimicking how people think instead of augmenting the process using computer technology. While this approach has made huge gains – especially in the fields of image recognition and speech recognition where machines are now more effective than humans in many publicly available benchmarks, its ability to cope with unexpected scenarios in the real world has been limited. As the true value of AI lies in how it can increase productivity in the real world, this limitation might prevent the wide adaptation of AI systems if not well-addressed.
But now a new approach is taking shape: human-centered AI. This keeps the end user at the forefront of the experience, and instead of trying to outperform humans, works with them to make them better at the task at hand. When humans and machines work together like this, the gains for human progress are much greater than AI going it alone.
Lacking a Bedside Manner: Where Modern AI Gets It Wrong
Currently, if you give an AI a concrete goal and a lot of data on how to achieve it, it can generally outperform the average performance of a human. It’s a simple formula: big data plus deep learning equals superhuman performance. While this undoubtedly captures headlines (think of the news coverage when IBM’s Watson computer won the gameshow Jeopardy in 2011), the worse case would be that performance of AI under unexpected scenarios in the real world is inferior than humans who can typically handle outliers gracefully while not under pressure or fatigue.
For example, autonomous vehicles are safer than human drivers on average performance, because they eliminate human errors under fatigue or distraction. However, their behavior in unexpected scenarios such as a flying balloon in front of the vehicle can be very dangerous.
Whereas a human might be surprised but can typically continue driving through such scenario safely. Another good example is in healthcare, when an AI system might be able to identify a tumor more accurately than a doctor on average since AI has observed more examples than each individual doctor. However, a human still needs to take the responsibility of the diagnosis and wins the patient’s trust by nicely explaining it.
This is where the human factor becomes crucial.
Man Vs Machine
Instead of removing people from the process, human-centered AI works with humans in order to empower and facilitate them. Stanford University’s Institute for Human-Centered AI is at the forefront of this approach. It has identified three areas of focus: to study and forecast the impact of AI on humans and society; how to build AI systems to facilitate humans; and to try to gain a deeper understanding of human intelligence in order to develop new AI technology.
This last one is vital. The current extent of machine learning plus big data doesn’t equate exactly to human intelligence. A human-centered approach moves AI beyond just data and algorithms to real-world applications with tangible societal benefits.
Humans are currently significantly better than machines at certain tasks, so it would be ridiculous to outsource every job to a robot. Humans are very good at learning from very few examples. When faced with a new scenario, before we can collect a lot of data, the best thing to do is to let humans gracefully handle and explore it. In these circumstances, human intelligence is much more efficient than AI.
Humans have a lot of sophisticated sensors too, and while some digital sensors are pretty good (such as a camera), robots can’t replicate a human’s sense of smell, taste or touch to anywhere near the same degree of sensitivity under an affordable budget. For example, a masseuse can tell where you’re sore just from touching you, but a machine without a good-enough touch sensor won’t be able to develop that kind of intuition. It wouldn’t be a very sophisticated massage.
We Can Rebuild it: The Key Elements of a Human-Centered AI
With any human-centered AI system, trust is key. To inculcate trust, you need to let the user know under what conditions the system should work perfectly, and when they should question its decisions. It’s also vital to temper its behavior. For example, if a self-driving car is updated with more data and suddenly starts driving much slower because the data says it’s safer, the user will think something is wrong, because it hasn’t met their usual expectations. If the behavior needs to change, it should do so gradually, just as people gradually change their own behavior as they learn.
Any areas with a very strong human-facing component will benefit the most from human-centered AI. For instance, in call centers, it was thought chatbots would be able to replace human agents, but even the best conversational AI still needs to be monitored by humans in case of errors. Healthcare is also a big one. Instead of replacing doctors and nurses with robots, human-centered AI can take over the diagnostic work (as long as it’s checked by a doctor) which would increase the number of patients a doctor can see in a day. It would also reduce the rate of diagnostic error. It makes doctors more efficient and effective.
It all comes down to identifying the user’s pain points and making sure the system addresses them. It’s also very important that an AI can explain its decisions, especially in a sector like healthcare. This will let the doctor see why it reached the conclusion it did, check whether it is correct, and explain it to the patient.
Human-centered AI is guiding us to develop AI in the right way. By focusing on addressing the end user’s pain points, we create an environment in which humans and AI can co-exist, with increased productivity, better standards of living and more benefits for society as a whole.
* This article is also appeared on The Business Times.
WE ARE HERE TO HELP
YOU MIGHT ALSO LIKE
From the day they were introduced on Netscape in 1994, cookies were intended to help companies enhance and personalize the user experience on the internet, whether it is authenticating logins or remembering language settings. Marketers have also depended on third-party cookies for more than two decades to track consumer behavior online, using cookies for targeting, retargeting, display advertising and behavioral marketing. However, with Google announcing that it will phase out the use of third-party cookies in Chrome by the end of 2022 – following similar moves by Mozilla’s Firefox and Apple’s Safari – the death of the third-party cookie is shaking up the digital marketing industry. The Growing Demand for Privacy While third-party tracking cookies is useful for optimizing online experience, there has been a growing concern over how consumer data is tracked, stored and shared. Consumers are often unaware how their behaviors are being tracked and used to market to them. Therefore, such tracking is seen as a breach of consumer privacy and trust as they might not have provided explicit consent to leverage their personal data. Moreover, consumers have no visibility into which companies are processing their data, beyond those that own the websites they visit. The
Q: What is GDPR? The General Data Protection Regulation (“GDPR”) is a new legislation by the EU parliament that lays out requirements for data collection, storage, and usage practice. This new law is meant to replace the 1995 EU Data Protection Directive (DPD) to significantly enhance the protection of the personal data of EU citizens and increase the obligations on organisations who collect or process personal data. Q: When is the GDPR coming into effect? The GDPR will become fully enforceable on May 25, 2018. Q: Who does the GDPR affect? Although the GDPR is an EU regulation, the territorial scope of GDPR is potentially far wider as it can also apply to non-EU businesses in certain cases. Businesses that market their products to or monitor the behavior of people in the EU are required to be GDPR compliant. Q: What constitutes personal data under the GDPR? Any information related to a natural person, defined as ‘Data Subject’ in GDPR, that can be used to directly or indirectly identify the person. It can be anything from a name, a photo, an email address, bank details, posts on social networking websites, medical information, or a computer IP address. Q. How is
Demand for e-learning has grown significantly over the past decade as people and companies increasingly look to upskill online. According to Business Wire, the global corporate e-learning market is expected to be worth US$49.87 billion by 2026. The US and Europe make up over 70 percent of this, while Asia Pacific is the fastest-growing region, with an anticipated growth rate of 20 percent a year. Some of the big players in the market include Udemy, SkillShare, and Coursera. Despite the positive prospect, what do e-learning providers need to do to play the long game in this thriving market? The answer lies in smart strategy and advanced technology like artificial intelligence (AI) and deep learning. The Rise of E-Learning This growth in the e-learning market is being driven by several factors: A glut of new e-learning technology, the need for talent enhancement and retention, and a desire for more convenient and cost-efficient ways to learn. The COVID-19 pandemic has only accelerated this growth by pushing people to spend more time online. However, competition in the e-learning industry is increasingly fierce, and there are still several challenges providers need to overcome. The mental hurdle is one. E-learning is still a relatively