The Fourth Industrial Revolution just took a huge step forward, thanks to a breakthrough artificial intelligence (AI) model that can learn virtually anything about the world — and produce the content to tell us about it.
The AI program is GPT-3 by OpenAI, which started out as a language model to predict the next word in a sentence and has vastly exceeded that capability. Now, drawing from voluminous data — essentially all of Wikipedia, links from Reddit, and other Internet content — GPT-3 has shown it can also compose text that is virtually indistinguishable from human-generated content.
getty
Asger Alstrup Palm, Area9’s chief technology officer, explained that GPT-3 was tasked with testing the “scaling hypothesis” — to see if a bigger model with ever-increasing amounts of information would lead to better performance. Although it’s too early to call the scaling hypothesis proven, there are some strong indications that this is, indeed, the case.
Further validating the potential of GPT-3, Microsoft recently announced it will exclusively license the model from OpenAI, with the intention of developing and delivering AI solutions for customers and creating new solutions using natural language generation. The takeaway for business leaders is that advanced AI models will make it possible to capture greater productivity improvements and training efficiencies — but only if leaders embrace, not fear, this technology.
Learning Everything about the World
GPT-3 stands out from previous iterations because it is the first AI model to exhibit meta-learning. What makes GPT-3 such a good “learner” is that, unlike earlier AI models, it does not require supervision. GPT-3 is able to absorb what it knows from multiple sources; the more it learns, the better it understands the world.
This model has demonstrated numerous capabilities: writing poetry and jokes, mimicking the literary styles of Ernest Hemingway and Jane Austen, summarizing movies, and more. It was trained for language, but somehow learned arithmetic and composing music for guitar — all without any new algorithms. When there is a glitch or an error, providing a different prompt (i.e., asking a different way) typically produces the correct answer.
It’s not perfect. Sam Altman, CEO of OpenAI, cautioned in a tweet, “The GPT-3 hype is way too much. It’s impressive…but it still has serious weaknesses and sometimes makes very silly mistakes. AI is going to change the world, but GPT-3 is just a very early glimpse.”
That glimpse, though, has been very impactful. As I learned about GPT-3 and its capabilities, I felt like I did the first time I saw a personal computer.
Commercializing Advanced AI
Next-generation AI will likely increase human productivity through collaboration with this technology. An example is computer programming. AI could be used to produce the first draft of computer code that programmers then customize or improve. Similarly, AI-produced written material could be generated quickly, with human “editors” to ensure the message and style, and add content where needed. It also works the other way around, with AI correcting grammar and spelling for humans.
This is game-changing, especially for learning. If corporate learning expected to have another 10 to 15 years before needing to adjust to widespread AI applications, the breakthroughs demonstrated by GPT-3 have shortened that timeframe to between 3 and 5 years. Change is coming in what and how people learn.
Five Ways GPT-3 Will Likely Change Learning
1. Knowledge on demand. As advanced AI models become vast repositories of information, learners at every level will have access to contextualized knowledge on demand. This will make rote memorization less important for humans. In fact, using an advanced AI model that can explain anything will become no different than using a calculator to find the square root of a number. Learners will need to acquire the skills to ask AI models for information and interpret results that will reflect the biases found on the internet. It will be much like Googling for information today. However, as AI models provide access to vast knowledge and get better in giving summaries and explanations, education at every level will likely shift away from knowledge acquisition and put more emphasis on knowledge application. To be clear, knowledge will still be important, particularly when people need to act quickly, instinctively, and competently. Indeed, the essential knowledge to carry out a person’s job must become second nature to achieve what we call automaticity. But the time and energy invested in gaining peripheral knowledge will be better spent in learning to apply that knowledge and solving problems.
2. 21st century learning will accelerate. As broad knowledge becomes more of a commodity that’s easily accessed, the true differentiator will be how skilled people are in applying that knowledge. For humans, this elevates the importance of 21st century skills, as identified by researcher Charles Fadel: communication, collaboration, creativity, and critical thinking. In fact, the competitive edge for organizations will likely be found in how well employees perform when using 21st century skills to innovate, solve problems, and engage in higher-order thinking. Projects, hands-on experiences, and other forms of collaboration to apply knowledge will be at the center of 21st century learning.
3. Workers get help with routine tasks. In the future, workers could get access to an AI-powered system that explains how to carry out routine tasks. This will not only save time and money for basic training; it could also help reduce errors by giving reinforcement-on-demand in how to perform one’s job.
4. Learners will get a personal learning coach. AI models that can digest and summarize information could become the ideal coach for learners to explain anything in the standard curriculum (from K-12 through corporate learning). In addition, AI models could come up with questions to help learners test their understanding.
5. More content means better learning. Education systems change slowly, even if technology makes part of the curriculum obsolete. However, over time, as APIs and other plugins from advanced AI become available, it will be easier (and less costly) to develop quality learning content. This, in turn, should increase demand for great content to facilitate and support better learning.
As we look to the future, learning must evolve. For corporate learning in particular, the priority must be ensuring that people are better equipped to collaborate, communicate, exercise creativity, and engage in critical thinking. Developing and deploying these skills will increase the unique value of humans in the 21st century and beyond.
