We cannot decide whether AI will replace humanity as the dominant intellectual force or whether it will just be a sideshow that never keeps its promise. Is AI the technology of the future or will it always be that way?
The answer is probably somewhere in between.
Bill Gates once stated that transformative technologies are overestimated in the short term and underestimated in the long term. AI is certainly one such transformative technology.
Last week, the New York Times published a cautionary story about how IBM tried to use its heavily branded AI solution called Watson to transform the business world and cement the future of IBM. The Times noted that IBM stock has fallen 10 percent in the years since Watson’s famous Jeopardy win over Ken Jennings, while stocks of AI / cloud competitors like Amazon and Google have penetrated the stratosphere.
Changing the world with a single product is more difficult than management at IBM expected. The Times wrote, “The company’s top management, as current and former IBM insiders noted, was until recently dominated by service and sales executives rather than technology product experts. Product people, they say, might have better understood that Watson was designed specifically for a quiz show, a powerful but limited technology. ”A tool made specifically for one job cannot always be easily converted for other jobs.
Currently, our only AI tools fall into the “narrow AI” category – tools designed to perform a function or limited groups of functions in a particular space. Siri may seem like she knows everything, but she doesn’t. It is designed to interpret speech in specific languages, access databases and Apple’s platform capabilities, and respond appropriately in the (admittedly impressive) limited circumstances it is likely to encounter. AI can shock us with speed and apparent breadth of knowledge in selected tasks that go far beyond what humans are capable of, but only for a certain series of actions.
AI can shock us with speed and apparent breadth of knowledge in selected tasks that go far beyond what humans are capable of, but only for a certain series of actions.
This was part of the IBM management problem. They knew that Watson was “smart” – she could beat two Jeopardy champions – and that such skills could easily be used to solve the big problems in medicine, finance, and government. Perhaps one day an AI will be able to do this, but it will solve these problems if it is designed to solve these problems. IBM made an excellent slim AI solution with Watson, but IBM assumed that there was general AI – one that could solve any problem. General AI only exists in science fiction and probably won’t be in our world in the near future.
If IBM is right about Watson, and Watson is likely to be a foundation for IBM’s business for the next 50 years, is that IBM has developed a business solution when IBM can use Watson-like AI to address a number of specific business problems solve that can be resold indefinitely to solve similar types of problems. If you replace a complicated business process for one bank, you can repeat the process for all other high margin banks. And if the business process is general enough, your solution can potentially be rolled out to many other types of businesses and government agencies. This is the Microsoft model, except with AI and not old-fashioned software. There are significant profits.
However, the problems being addressed must surely be within the capabilities of the AI, which will likely involve grouping, including or excluding samples from the groups, and applying logic to huge data sets. Michael Jordan proved to himself and the world that while he was the greatest basketball player of his day, he couldn’t simply transfer this athleticism to other sports. We have never seen a person who is the best athlete in all sports, the brightest scientist, the greatest public speaker, and the most victorious jeopardy champion. Why does IBM expect an AI equivalent to be able to do it? Why should it be?
We have never seen a person who is the best athlete in all sports, the brightest scientist, the greatest public speaker, and the most victorious jeopardy champion. Why does IBM expect an AI equivalent to be able to do it? Why should it be?
The new management of IBM recognized this truth and embraced it to a profitable extent. The Times articles stated: “Watson is now a collection of software tools that companies use to build AI-based applications – those that streamline and automate primarily basic tasks in areas such as accounting, payments, technology operations, marketing, and customer service . It’s an artificial intelligence workhorse, and so is most of the AI in today’s business world. . . IBM says it has 40,000 Watson customers in 20 industries worldwide, more than twice as many as four years ago. ”Watson may still be the foundation for IBM’s future business, but it will do so by using its limited strength on specific problems.
A recent McKinsey report suggests that around half of current work activities can be automated using AI and other technologies. This includes data collection and processing as well as physical tasks in structured environments. It is therefore likely that AI, which is already transforming the workplace, will be seen as a major force in business and government efficiency in the near future. And the education and maturity that comes with learning the strengths of this technology will enable AI to evolve to the next level.
Will we ever produce general AI that accepts input in dozens of different areas and provides answers to our deepest problems? For better or for worse, we won’t have this technology anytime soon. But it is unrealistic to expect. Watson is a spectacular technology to meet the challenges for which it was designed. We need to recognize its potential productive uses and not get lost in dreams of a panacea. Because, as science fiction tells us, general AI can create its own problems.
Copyright © 2021 Womble Bond Dickinson (US) LLP All rights reserved.National Law Review, Volume XI, Number 201