This article is part of our series that explores the artificial intelligence business.
Digital technologies, and at the forefront of their artificial intelligence, are driving fundamental changes in society, politics, education, economics and other fundamental aspects of life. These changes provide opportunities for unprecedented growth across different sectors of the economy. But at the same time, they bring with them challenges that organizations must overcome before they can reach their full potential.
In a recent speech at an online conference hosted by Stanford Human-Centered Artificial Intelligence (HAI), Stanford Professor Erik Brynjolfsson discussed some of these opportunities and challenges.
Brynjolfsson, who heads Stanford’s Digital Economics Laboratory, believes that in the next decade, the use of artificial intelligence will be much more widespread than it is today. But its adoption will also face a period of pause, also known as the J-curve.
“There’s a growing gap between what technology is capable of and what it’s already doing, as opposed to how we respond to it,” says Brynjolfsson. “And there lie many of our society’s greatest challenges and problems and some of our greatest opportunities.”
Machine learning and higher productivity
According to Brynjolfsson, the next decade will see significantly higher productivity thanks to a wave of powerful technologies — especially machine learning– which find their way into every computer device and application.
Progress in computer vision were huge, especially in places such as image recognition and medical imaging. Talking to phones, watches and smart speakers has become commonplace thanks to advances in natural language processing and speech recognition. Product recommendation advertising placementinsurance signing, loan approval, and many other applications have benefited greatly from advances in machine learning.
In many places, machine learning reduces costs and accelerates production. For example, the application of large language models in programming can help developers become much more productive and achieve more in less time.
In other areas, machine learning can help create applications that did not previously exist. For example, generic deep learning models create new applications for arts, music, and other creative work. In areas such as online shopping, advances in machine learning can be created major changes in business modelssuch as moving from “shopping-then-shipping” to “shipping-then-buying.”
The confinements and urgency caused by the covid-19 pandemic have accelerated the adoption of these technologies in different sectors, including remote tools, robotic process automation, powered drug research, and factory automation.
“The pandemic has been terrible in many ways, but another thing it has done is that it has accelerated the digitalization of the economy, compressing in about 20 weeks what would have taken perhaps 20 years of digitization,” Brynjolfsson says. “We have all invested in technologies that enable us to adapt to a more digital world. We will not stay as far away as we are now, but we will not return all the way. And that increased digitization of business processes and capabilities is squeezing the deadline for us to adopt these new ways of working and ultimately lead to higher productivity. “
The J-curve
The productivity potential of machine learning technologies has one big caveat.
“Historically, when these new technologies become available, they do not immediately translate into productivity. There is often a period where productivity is declining, where there is a break, ”says Brynjolfsson. “And the reason this break is is that you have to reinvent your organizations, you have to develop new business processes.”
Brynjolfsson calls this the “Productivity J-Curve” and documented it in a paper published in the American Economic Journal: Macroeconomics. Basically, the great potential caused by new general-purpose technologies such as the steam engine, electricity, and more recently machine learning requires fundamental changes in business processes and workflows, the co-invention of new products and business models, and investment in human capital.
These investments and changes often last for several years, and during this period they do not give perceptible results. During this phase, the companies create “intangible assets,” according to Brynjolfsson. For example, they may be training and retraining their workforce to use these new technologies. They may be restructuring their factories or instrumenting them with new sensor technologies to take advantage of machine learning models. They may need to upgrade their data infrastructure and create data lakes on which they can train and operate ML models.
These efforts may cost millions of dollars (or billions in the case of large corporations) and will not make a difference in the company’s production in the short term. At first glance, it seems that costs increase without any return on investment. When these changes reach their turning point, they result in a sudden increase in productivity.
AI J curve
“We’re in this period right now where we’re doing a lot of that painful transition, restructuring work, and there are a lot of companies that are struggling with that,” says Brynjolfsson. “But we’re working on that, and these J-curves will lead to higher productivity — according to our research, we’re close to the bottom and turning up.”
Making the transition to AI
Unfortunately, adapting to AI and other new digital technologies is not working in a predictable way. Most companies do not make the transition right or lack the creativity and understanding to make the transition. Various studies show this most applied machine learning projects fail.
“Only about the top 10-15 percent of firms make the most of the investment in these intangibles. The other 85-90 percent of firms are behind and hardly need any of these restructurings,” says Brynjolfsson. “This is not just the big tech companies. This is within every industry, manufacturing, retail, finance, resources. In every category, we are seeing the major companies withdraw from the rest. There is a growing performance gap.”
But while adopting new technologies will be difficult, it is happening much faster compared to previous cycles of technological advances because we are better prepared to make the transition.
“I think it’s obvious that it’s going to happen a lot faster in part because we have a much more professional class of people trying to study what works and what doesn’t,” Brynjolfsson says. “Some of them are in business schools and academies. Many of them are in consulting firms. Some of them are journalists. And there are people who describe which practices work and which don’t.”
Another element that can help a lot is the availability of machine learning and data science tools to process and study the huge amount of data available about organizations, people and economy.
For example, Brynjolfsson and his colleagues are working on a large database of 200 million jobs, which includes the full text of the job description along with other information. Using different machine learning models and natural language processing techniques, they can transform the jobs into numerical vectors, which can then be used for a variety of tasks.
“We think of all jobs as this math space. We can understand how they can relate to each other, ”Brynjolfsson says.
For example, they may make simple conclusions about how similar or different two or more jobs are based on their textual descriptions. They can use other techniques such as clustering and graphically neural networks draw more important conclusions as to what skills are most in demand, or how the characteristics of a job would change if you modified the description to add AI skills such as Python or TensorFlow. Companies can use these models to find gaps in their employment strategies or to analyze the employment decisions of their competitors and leading organizations.
“Such tools just didn’t exist as recently as they did five years ago, and I think it’s a revolution as important as the microscope or some of the other revolutions in science,” says Brynjolfsson. “We now have them for social science and business to have such visibility. That allows us to make a transition much faster than before.”
However, Brynjolfsson warns that not many companies use such tools. This may be further evidence from his previous point that companies have not yet figured out the right transition strategy and are relying on old methods to restructure and adapt to the age of AI. And at the heart of this strategy should be the proper use of human capital.
“You have hundreds of billions of dollars of human capital, from skills coming out the door, and then the company is trying to hire back people with the skills they need. What they don’t realize is that the workers they let out often had skills. , who were very close to those they hired, ”Brynjolfsson says.
With the help of machine learning, they will have better visibility and knowledge of their “capable neighbors,” says Brynjolfsson. For example, a company may find that instead of hiring a lot of people and looking to hire new talent, maybe all they have to do is retrain a little and reuse their workforce.
“It’s much more expensive to hire someone fresh than it would have been for them to take some of those people who are already in the company and say, if we teach you Python or customer service skills or other skills, you can do this job that we’re looking for “My hope is that in the next decade, workers will be in a much better position to take full advantage of their skills and abilities. And it will also be good for companies to understand all the value they have there. and machine learning can help a lot in understanding those relationships. “
This article was originally published by Ben Dickson on TechTalks, a publication that examines trends in technology, how they affect the way we live and do business, and the problems they solve. But we also discuss the downside of technology, the darker implications of new technology, and what we need to pay attention to. You can read the original article here.
Source
Ben Dickson