Artificial intelligence and ChatGPT unpacked
But first let's set the stage
Artificial intelligence is a broad term that refers to computer software which engages in activities we usually regard as human, such as learning, planning and problem-solving. It’s capable of making decisions on its own and is, well, very smart.
There are many forms of AI already commercially available and in daily use, aimed at making everything easier. Examples include manufacturing robots, virtual booking systems, self-driving cars, those dreaded customer chat-bots, robot vacuums, digital assistants (Siri, Alexa et al) etc. Basically, it’s everywhere and increasing in prevalence.
What it’s not very good at, is creativity. But wait, this is where generative AI comes into play.
Generative AI is a machine learning subfield that uses algorithms to generate new data. It takes existing data and uses this data to create new content that hasn’t existed before in the non-digital world. Examples include writing, designing, coding as well as creating audio, video and static images.
Now ChatGPT is an example of a generative language model. It uses huge datasets of existing text data to understand how human writing and speech patterns work. Then, given a prompt, it uses a transformer algorithm to generate its own text in response.
So it generates human-like responses to questions and prompts.
And get this, it can adjust and change the tone and messaging of the text on demand. For example, you can request this software to write an article for you on the benefits or otherwise of working from home. Once you’ve reviewed the article, you can instruct it to lengthen or shorten it, make the tone more formal or informal, add or delete facts etc. And so on.
The possibilities are endless.
Trying out ChatGPT
Time to try this out, I thought. So, I gave it the following prompt: “ways in which AI will affect corporate learning and development”.
In under five minutes it gave me the following offering:
AI has the potential to greatly impact corporate learning and development in several ways:
Personalized learning: AI-powered systems can analyze an individual's performance, preferences and prior knowledge to create a personalized learning experience. For example, an AI-powered system can track an employee's progress through a training program, and adapt the content and pace of the program to match their needs. This can increase engagement and retention of the material and make the learning experience more effective.
Automated content creation: AI can be used to analyze existing content and automatically generate new learning materials, such as quizzes, summaries, and flashcards. For example, an AI-powered system can read through a company's internal documents and create a set of multiple-choice questions to test employees' understanding of the information.
Learning analytics: AI can be used to analyze data on how employees are interacting with learning materials, providing insights into what is working and what needs to be improved. For example, an AI-powered system can track an employee's progress through a training program and identify areas where they are struggling. The system can then provide additional resources or support to help the employee improve their understanding of the material.
Intelligent tutoring: AI-powered systems can provide real-time feedback and support to learners, helping them to stay on track and achieve their goals. For example, an AI-powered system can provide instant feedback on an employee's performance during an online training program, and offer suggestions for improvement.
Automation of administrative tasks: AI can automate repetitive and time-consuming tasks, such as scheduling, tracking progress, and assessing learning outcomes, freeing up more time for human employees to focus on more strategic tasks. For example, an AI-powered system can automatically schedule and track employee training, and generate reports on the progress and completion of training programs.
Skill-based learning: AI can help companies identify the necessary skills for their workforce and tailor the learning and development programs accordingly, in order to meet the ever-changing demands of the market. For example, an AI-powered system can analyse the job market and identify the skills that are in high demand, and then suggest training programs to help employees acquire those skills.
Mmmm….. Not a bad list at all! That’s pretty neat. Admittedly I asked the program to add examples, which I promptly got.
But what can't ChatGPT do?
I’d already made my own list of what I thought were essentially human elements that are hard to replace with artificial intelligence, no matter how smart the program is. My list included elements like creativity, empathetic learner interaction and support, contextual understanding, emotionally intelligent responses, flexibility to change direction, and integration with the bigger picture.
But hey, let’s ask our AI friend what they come up with.
And here are the results.
AI has limitations in corporate training and development, such as:
Lack of creativity: AI systems can perform tasks based on pre-programmed instructions, but they lack the ability to think creatively and come up with new ideas.
Limited understanding of context: AI systems may not fully understand the context in which a task is being performed, which can lead to errors or mistakes.
Difficulty with open-ended tasks: AI systems may struggle with tasks that are not well-defined or that require a high degree of flexibility.
Limited emotional intelligence: AI systems may not be able to understand or respond to emotional cues, which can be important in training and development.
Lack of human interaction: AI-based training may lack the personal touch and human interaction that can be important for learning and engagement.
Bias: AI algorithms can perpetuate bias if the data it is trained on is biased, which can lead to unfair or inaccurate results.
Lack of critical thinking: AI systems are not able to engage in critical thinking and reflection, which are important skills for professional development.
Again, pretty neat! This AI displays a whole lot more insight into its own limitations than most people 😉.
So, what does this mean for corporate training? Will ChatGPT take everyone's jobs?
Here’s what ChatGPT told me:
It is unlikely that this or any other AI technology will take everyone’s jobs. While AI and automation can certainly perform some tasks more efficiently than humans, there are many tasks that require human judgement, creativity and empathy that AI systems are not capable of. Additionally, as AI technology advances, it is likely that new jobs will be created that require skills that are not easily automated. It is important for individuals to continue to learn and adapt to new technologies in order to remain competitive in the job market.
Human judgment. Creativity. Empathy. Yes, these are uniquely human attributes that when it comes to human judgment, we need to view ChatGPT, like other AI tools as an enabler. Sure, it’s a very smart technology-driven enabler, but at the end of the day it assists. It doesn’t create the vision or pull different and complex strands of subjective and objective input together.
And who decides what’s best in a given situation? The human, that’s who. Matching training solutions to organizational context requires complex analysis and decision-making to get the best result.
And when it comes to creativity, the same applies.
AI software exists which creates something that resembles art – but there’s a philosophical debate raging whether non-human creations can be regarded as art, so let’s not go there today. But let’s agree that creativity uses imagination – and that imagination is a human characteristic.
Creativity in corporate learning is what happens when we find new and unique ways to present ideas and concepts. It’s what happens when something we learn sparks further ideas in us.
And then we move to implement those ideas to benefit us and those around us.
And then of course there’s empathy. Artificial intelligence can simulate feelings, but it can’t feel. And that’s not genuine. And we all know the difference.
When a chatbot “cares about you and your customer experience” we all know what that means. An eyeroll from us because we know that care is what we get from people with whom we have a genuine connect and a deep relationship.
And empathy is what we experience when some-one puts themselves in our shoes and treats us with support and understanding.
An empathetic instructor, mentor, trainer or facilitator can dramatically accelerate how we receive and apply learning. Because we feel they get us. And they care about our learning experience.
Corporate training’s about to get a whole lot more interesting.
But it’s good to confirm that the things that make training and development human will remain human. Because after all we’re focusing on developing humans.
The machines will develop themselves in their own race to technological excellence.