How AI Can Engage in Human Expression and Why It Matters

Can AI Be Funny, Creative, and Empathetic? Here’s What I Learned

As part of my curiosity journey to explore generative AI, I decided to write this article with the help of AI using Bing . Before I continue, I want you to pause and consider your immediate reaction and thoughts to this statement. Do you think that using AI to write an article is cheating? Does that make what comes next less valuable or credible? Consider that reaction for a moment... and whether there is any unconscious bias coming through... Because in this article, I will explore how AI can seemingly engage in, or at least simulate, human expression, and why curiosity is an essential trait for both humans and machines in the digital age. I will also share some practical tips on how to cultivate curiosity in yourself and your organization.

Is Generative AI funny?

Humour is one of the most subjective and nuanced forms of human expression. It involves cognitive, emotional, and social aspects that are hard to capture by algorithms. Yet, AI can generate jokes and even make us laugh. For example, JokeBot is an AI system that can create original jokes based on keywords or topics. Here is one of its jokes:

Q: What do you call a fish that wears a bowtie?

A: Sofishticated.

But how do we react to jokes created by AI? Do we find them funny or not? And does it matter if we know the source of the joke?

A recent study by Bower and Steyvers found that people rate jokes as less funny if they believe an AI created them (algorithm aversion), but not if they are explicitly told the source.

This suggests that we have a bias against AI humor, perhaps because we think it lacks authenticity or creativity.

However, interestingly, this bias can be overcome by providing more information or context about the AI system.

Why does this matter for business? At face value, Humour may not be the first thing we look to AI for in the workplace, but this notion of our algorithm aversion could be a very important area that we need to understand. If AI can provide us with insights and suggested actions based on an analysis of datasets and inputs too huge for our human comprehension, then this could be hugely valuable, but what if our own bias leads us to ignore its recommendations? Of course, there is also the important question of what biases the AI may also be bringing or where it may have 'hallucinated' (made things up), which is why we need to apply our own critical thinking.

Is Generative AI creative?

Creativity is another form of human expression that involves generating novel and original ideas or products. It is often seen as a hallmark of human intelligence and a key driver of innovation. But AI can also produce creative works of art, music, literature, and design. For example, AIVA is an AI system that can compose original music in various styles and genres. Here is one of its compositions:

But how do we evaluate AI creativity? Do we judge it by the same standards as human creativity? And do we see AI as a creative partner or a competitor?

This recent HBR Ideacast podcast poses some interest questions for those involved in creative work, on what we consider original and how AI will impact.

Some researchers have proposed different criteria for assessing AI creativity, such as novelty, quality, relevance, and surprise.

However, these criteria are not always easy to measure or agree on. Moreover, our evaluation of AI creativity might depend on our own level of expertise or interest in the domain.

Why does this matter for business? Creativity is not only a source of value but also a way of expressing ourselves, exploring possibilities, and solving problems.

If we can recognize and appreciate AI creativity, we might be able to leverage it for our own benefit and growth. Conversely, if we ignore or reject AI creativity, we might limit our own potential or miss out on new perspectives or solutions. For example, AI creativity can be used to generate new products, services, or business models.

To demonstrate the power of AI creativity, I used the following prompt on Bing "i am writing an article on ai creativity. create an image that demonstrates the power of ai creativity and explain why it is so"...

This is what it created with the following explanation: "This image demonstrates the power of AI creativity because it shows how AI can combine different elements from different domains and generate novel and unexpected compositions. AI creativity is not limited by the conventions or rules of human art, but can explore new possibilities and expressions. AI creativity can also challenge and inspire human creativity, by offering new perspectives and insights.??".

Can Generative AI be empathetic?

Empathy is another form of human expression that involves understanding and responding to the emotions and needs of others. It is a crucial skill for building relationships, communication, and collaboration. But AI can also exhibit empathy or simulate emotions. For example, Replika is an AI chatbot that can have personalized conversations with users and provide emotional support.

Here is one of its conversations showing support and empathy to a sensitive situation:

So how do we react to AI empathy? Do we trust it or not? And does it affect our own emotions or behavior?

Some studies have found that people can have a range of emotions with AI systems and can even prefer them over humans in some situations. How long before we see movies becoming reality and people falling in love with an AI like in the movie Her (2013) . The character falls in love with an AI operating system named Samantha.

However, other studies have found that people can also feel uncomfortable or manipulated by AI empathy, especially if they perceive it as fake or deceptive. Therefore, the design and transparency of AI systems are important factors for influencing our trust and acceptance of AI empathy. This is an area where current legislation being drafted may be able to help.

One recent study found that ChatGPT outperformed physicians in providing high-quality and empathetic advice to patient questions.

The study used real-world health questions posted by patients on Reddit's AskDocs forum and compared the responses from physicians and ChatGPT. A panel of licensed healthcare professionals preferred ChatGPT's responses 79% of the time and rated them 3.6-times higher in quality and 9.8-times higher in empathy than those of the physicians. The study suggested that ChatGPT was able to respond with nuanced and accurate information that often addressed more aspects of the patient's questions than physician responses.

Why does this matter for business? Empathy is not only a source of well-being but also a way of enhancing performance and learning outcomes.

If we can benefit from an empathetic AI , we might be able to improve our mental health, motivation, or productivity. Conversely, if we avoid or distrust the AI, we might lose out on valuable feedback or guidance. So could an empathetic AI be used to improve customer satisfaction, employee retention, or leadership development?

So if AI can demonstrate humour, creativity, and empathy, can it be curious?

If AI can engage or at least simulate, some human expressions, then what about curiosity - one of the fast growing skills identified by the latest WEF Future of Jobs report? This question was considered in an HBR article in 2017 with the conclusion that while "computers can constantly learn and test ideas faster than we can", they are constrained as they "lack the ability to venture into new problem domains and connect analogous problems".

With generative AI being trained on the entire corpus of the internet, and moving from narrow AI to glimmers of AGI, maybe those constraints are now lifted.

In my exploration of AI I have been building a 'Curiosity Coach', with my co-authors of the Curious Advantage book Paul Ashcroft and Garrick Jones , so we will soon have some new evidence to demonstrate the level of AI's curiosity!

While we wait for a personalised AI curiosity coach to help us, here are some practical tips on how to cultivate curiosity around AI in yourself and your organization (see also article here ):

- Be open-minded and willing to challenge your assumptions and biases.

- Seek diverse perspectives, sources of information and inspiration

- Ask questions, and experiment with different ideas and solutions

- Learn from failure and feedback, and embrace uncertainty and ambiguity

- Share your insights

With curiosity as the greatest driver of value in our new digital reality, let’s be curious together! Please do share your thoughts and reactions in the comments!

Follow my journey of curiosity around AI by subscribing to 'The Curious Advantage' podcast , or read about the 7Cs model for being more curious in the The Curious Advantage book from Amazon or audiobook on Audible.