7 AI forudsigelser for 2020

Dr. Michael Feindt – founder of Blue Yonder and JDA’s chief scientific advisor – has been researching, understanding, and identifying ground-breaking use cases for artificial intelligence (AI) for many years. As a physicist and data scientist, he is constantly thinking about how algorithms can be developed, improved and advanced to not only benefit businesses, but improve our world.

AI and causality will converge to improve complex decision strategies.

Whereever complex policies should be improved by AI, learning from correlations in historical data is not enough – you need a causal understanding in order to make valid what-if statements. As an example, identifying casual shoppers and understanding which promotions will influence them most effectively, simply can’t be calculated by human brains. This is where AI and machine learning (ML) play a key role. Rather than targeting loyal customers, retailers should be focusing efforts on engaging more casual ones. Using causal AI-driven knowledge will make it possible to identify the right type of customer to target and what price to set. This will put retailers in a position to convert additional sales from casual shoppers and maximize margins on sales that would have been made by loyal shoppers anyway.

AI will drive the need for more data scientists, not less.

There’s a continued fear that AI will take away jobs given its ability to transcend human computational intelligence. The opposite is true. There will always be a need for humans to build, steer and monitor AI-based systems. There is a shortage of experts who think quantitatively and understand this level of mathematical algorithms. I equate this to when computers were invented. People thought the office worker function would die as a result. But 60+ years later, we still have too few experts that understand computer programming. We aren’t there with AI either, and won’t ever not need humans. AI doesn’t build itself or take control; that’s where humans come in.

AI will get a seat at the table.

This will be done via a new breed of leader that both understands AI and the business needs and challenges to drive the strategy. Data scientists can’t do it all; there needs to be an intermediary with the business acumen and technical AI knowledge that can help the company drive its initiatives forward. This leader will drive change management and trust at the executive level as that is still the biggest roadblock to AI adoption.

AI will drive sustainability.

Technology and sustainability go hand-in-hand. You have to have the means to be sustainable using technology, and AI technology is the only way to achieve this. By leveraging AI algorithms, companies can, measure environmental and social impacts, automatically make responsible corrections, and optimize operations for sustainability. Though the sustainability challenge grows more complex every day, these technologies can help businesses to operate responsibly — and profitably — via reduced waste, more efficient production, smarter transportation strategies, and reduced resource consumption.

Science education and curriculum will shift to data science.

There’s a movement for science overall to become more quantitative and data-driven. This includes the fields of medicine, energy, climate research, physics, chemistry, and even psychology. Scientists need to understand data science more than ever at the outset of their education, and there needs to be more public research made available on data science. The only way to do that is expanding the breadth of data science into curriculum for students.

AI will drive fairness and ultimately build a better, more diverse world.

Many AI systems can have the same prejudice as humans if they are trained on human decisions that are inherently biased. This is because AI will copy human behaviors as it learns. This was discovered initially by Google in its facial recognition software. The technology recognized Caucasian faces better than other races simply because that is what it was mainly trained on and therefore learned. Once discovered, researchers started to develop algorithms to make machine decisions fair even if the data it trained from was unfair. I believe these algorithms will ultimately be used as a way to fight discrimination and ultimately, build a better, more diverse world.

AI will move beyond the borders of one company.

When AI is integrated into the supply chain at a company, for example, it infuses intelligence and predictive capabilities that drive smarter, action-based decisions. If AI is integrated horizontally across the company and its suppliers, the power is in the ability to plan better on both sides. For example, if the supplier knows what the retailer will order and vice versa, you will avoid having too much safety stock and plan better on both sides. AI will ultimately optimize the supply chain ecosystem – not just one company. This is where I see AI really taking off in terms of supply chains, driving the power of these decisions globally and breaking down silos.

What do you think? What are some of the ways AI will continue to be a game-changer in 2020 and beyond? Tell us on social media! @JDASoftware

"Forbes: Sådan kan AI hjælpe detailhandleres beslutningstagning"

"Wall Street Journal: True Value Retools Hardware Supply Chain with JDA software solutions."