The Future of AI and Robotics Across IndustriesThe Future of AI in FinTechThe Future of AI in FinTech

In the 2020 documentary, “AlphaGo”, an AI-powered computer program of the same name developed by Google’s DeepMind research lab takes on Lee Sedol, world champion Go player, in a five-match tournament. Before the two face off in the ancient Chinese strategy board game, Sedol, who is certain he will emerge undefeated, says: “I believe that human intuition is still too advanced for AI to have caught up.”

It’s an evocative moment. Not just because a savvy audience can spot the narrative tension building, but because we really do root for Sedol to beat the machine. The human champ suffers three agonizing defeats before finally winning the fourth match—his only win—and when he does, it feels as though all of humanity breathes a sigh of relief.

It doesn’t escape the audience that behind the relatively simple computer program are dozens of deep learning and machine learning programmers and developers, which poses the question: Are we really rooting for a human versus a machine when the very program itself was built by humans? Or are we rooting for one man versus 20?

Alternatively, what happens when AI and robotics are tasked not with winning a game but with preserving human life?

In this article, we’ll look at some of the ways in which advancements in AI and robotics are creating opportunities, gaining efficiencies, and transforming manufacturing and health care, while also touching on the ethical uses of AI in areas like law enforcement.

Many aspects of robotics involve AI, since to perform precise or dangerous tasks, machines need to be equipped with at least a semblance of human senses to be able to react to external stimuli, whether vision, touch, or the ability to sense temperature.

The hope is that with the help of AI, surgical robots can assist doctors and surgeons and be useful in extraordinary times such as during a global pandemic, like COVID-19, so that potentially lethal exposure is minimized. As surgical robotics continue to evolve, “AI-enabled robots will eventually use computer vision to navigate to specific areas of the body while avoiding nerves and other obstacles. Some surgical robots may even be able to complete tasks autonomously, allowing surgeons to oversee procedures from a console.”¹

But before all that, the best AI application currently available in health care involves imaging. According to Rowan Smith, a patent attorney and partner at the firm Messner Reeves, “AI-based analysis of medical imaging data is poised to dramatically reduce costs of medical diagnosis leading to, hopefully, a massive expansion in access to preventative care and early disease detection—something that will lead to enormous savings in health care delivery.”

The automotive industry has long relied on people to perform most of the repetitive, exhausting, and dangerous tasks associated with mass production. Robotics and additive manufacturing (or 3D printing) have revolutionized that system, driving business efficiency and creating safer work environments. 

“I have no doubt that, because of advancements in AI and robotics, in 20 to 30 years, the automotive manufacturing process will have no human involvement at all,” says Smith. “Couple that with 3D printing of cabin components and completely autonomous automobile manufacturing appears to be inevitable.”

Beyond just automobiles, Smith believes that we will see a significant expansion of autonomous manufacturing: product manufacturing at the time an order is placed. “If drop shipping revolutionized product delivery and the web-based marketplace in the 2000s and 2010s, just-in-time manufacturing will characterize the 2020s and 2030s,” he adds.

Manufacturing a car or a couch via AI is one thing, but when it comes to deploying AI and robotics in law enforcement, an entirely different set of moral, legal, and ethical concerns arise. Police deployed robots can and have been used to take out an active shooter.

Sophisticated robots are being built around the world with little regulatory oversight, something which Elon Musk has been very vocal about warning us against. These robots will use AI for facial recognition, predicting the actions of people, or deciding to fire “nonlethal” projectiles. Many researchers find this problematic, because “… many of today’s algorithms are biased against people of color and others who are unlike the white, male, affluent, and able-bodied designers of most computer and robot systems.”²

This begs the question: How can we create algorithms that use unbiased judgment, when we ourselves suffer from overt and unwitting biases?

Adds Smith, “Although robotic or AI-driven law enforcement could bring benefits, many significant risks would still remain. AI systems as currently conceived are primarily pattern-matching systems and simply do not share human values. The risk that such an AI system makes a purely “logical” decision that directly contradicts a core human value is non-trivial. AI systems today commonly struggle with significant bias issues, and should those types of issues be present in an AI system configured to deliver lethal force, tragedies could easily result.”

Some of the most striking moments during Lee Sedol’s matches with AlphaGo came when we saw the human champion glancing up from the board at his opponent. But the person sitting across from Sedol wasn’t really his opponent — it was the man in charge of placing the pebbles on the board based on the computer’s desired move. In reality, Sedol was up against an invisible opponent, one who did not make eye contact, sigh, or rub his temple, and one who was impossible to read.

One of the most pertinent questions we will have to answer in our lifetime is: What will our relationship to AI look like?

Those learnings are just the tip of the iceberg. At Columbia Engineering, the belief that technology cannot exist without humanity is a core driving principle to building the frameworks for a healthy, connected, and creative world. Our online executive education AI program was developed to help business leaders create a vision for how AI can be used to transform services, build new products, optimize operational efficiency, and disrupt all facets of industry. The program gives learners both a 30-thousand-foot view and the deep technical expertise to lead engineers, developers, and programmers in executing their vision. As a learner in the AI executive education program, you will not only receive a holistic education capturing the fundamentals of AI, such as design and analysis of efficient algorithms, theoretical underpinnings, architecture, performance, datasets, and applications of neural networks and deep learning (DL); you’ll also explore how these topics relate to issues like security, privacy, data mining, and storage, as well as their legal and social contexts and frameworks.

¹Robotics in Healthcare to Improve Patient Outcomes. Accessed November 2021.arrow_upwardReturn to footnote reference

²Berreby, David. “Can We Make Our Robots Less Biased Than We Are?” New York Times, Nov. 2020. Accessed November 2021.arrow_upwardReturn to footnote reference