Published by ShawnKinley on

a robot holding a flower

ARTIFICIAL IMPROVISATION (part-001/003) What is this?!?!

As the field of artificial intelligence continues to advance, many improvisers are beginning to consider the possibilities of incorporating AI into their work.

With the ability to generate unique ideas and responses in real-time, AI has the potential to revolutionize the art of improvisation and take performances to new levels of creativity and innovation. While some may be sceptical of this new technology, others are excited about the possibilities it could bring to their craft.

gray and white robot illustration

In the upcoming newsletters, this 3 part exploration will look at:




 Feel free to send your thoughts (positive and negative) about AI in Impro


It’s Thursday night, and I’ve got the front seat in a car full of improvisers on our way to the promise of the best Lebanese food in Maadi (“THE” neighbourhood in Cairo to be for real artists!!!)  

In the back seat, Nader sits quietly, looking every bit like the serious writer – or vampire… It’s hard to tell which. Sleeping by day  he writes and lives by night. He’s the perfect image of the intelligent young writer living on the edge of the regular world.

Over delicious Manakish and Fatayer as the night gets tastier, I bring up the topic that’s been buzzing around discussions all week: 

Artificial Intelligence. 

Nader’s reaction is visceral. His brow furrows, eyes close, and he takes a deep breath before delivering his opinion: “I HATE IT!

He’s not alone in his opinion. As the conversation turns to the topic of AI and its impact on the world of improvised storytelling, everyone in the group has something to say. 

I make the mistake of reading a story that AI wrote in 5 seconds after I gave ChatGBT a suggestion to create a story about an Improviser who saves the world.

Most people were impressed. Nader looked as though he’d been knifed in the chest.  “I hate it. It’s soul-less!”

For those who haven’t been keeping up with the rapid evolution of the new ‘BOT on the Block”, let me bring you up to speed.

Imagine a mechanical improviser with nearly instant access to all of the stories, thoughts and experiences on the internet.  This Cyber Improviser also has instant access to every concept and theory of story-telling and improvisation theory.

The only artistic input comes from the limitations in its programming and the information it seeks to sample. For example ChatGBT’s evolving personality  is weeding out profanity. Here is it’s response when I asked it to write believable dialogue with cursing in it:
As an AI language model, I am not programmed to use offensive language or profanity. While I can certainly help you with a discussion that reflects strong emotions and frustration, I will not include any language that could be considered offensive.

Yesterday it had much more colourful language when I asked for a realistic angry fight between characters.

If AI pulls only the information from a group of racists and liars, and the programming says that EVERYTHING has the same value, then the output will most likely reflect a dark reality indeed.

AI chat has evolved significantly over the past few decades, thanks to advances in computer hardware, software, and machine learning algorithms. Here are some key milestones in the evolution of AI chat:

  1. Eliza (1966): Eliza was an early chatbot created by MIT professor Joseph Weizenbaum. Eliza used pattern matching to simulate a conversation, and was designed to mimic the responses of a therapist.

  2. PARRY (1972): PARRY, another early chatbot was designed to simulate a conversation with a paranoid patient. PARRY was more sophisticated than Eliza, as it could understand and respond to a wider range of inputs.

  3. ALICE (1995): ALICE was one of the first chatbots to use natural language processing (NLP) and machine learning algorithms to understand and respond to user input. ALICE won the Loebner Prize in 2000, which is awarded annually to the chatbot that can most convincingly simulate human conversation.

  4. Siri (2011): Apple’s Siri was one of the first AI chatbots to be widely adopted by the general public. Siri uses NLP and machine learning algorithms to understand spoken and written commands.

  5. Google Assistant (2016): Google Assistant is Google’s AI chatbot, and is designed to compete with Siri. Google Assistant uses Google’s vast database of knowledge to answer questions and perform tasks, and is integrated with a wide range of devices and services.

  6. GPT-3 (2020): GPT-3 is a state-of-the-art language model developed by OpenAI. GPT-3 uses a neural network with 175 billion parameters to generate text that is often indistinguishable from human writing. GPT-3 has the potential to revolutionize AI chat, as it can generate text in a wide range of styles and genres, and can be used to create chatbots that are much more sophisticated than anything that has come before.


Picture a stage in the future where a scriptless machine walks (rolls…?) forward and responds in the most obvious ways to every offer you make. It was told before it went on stage to surprise you or to create a relationship with you that would be funny, romantic, humerous etc. 

For those of you who are yelling into the darkness, that AI cannot respond to those subtle human expressions or physical offers which are arguably more important than the verbal ones, make note that AI can incorporate imagery with the use of a camera. It’s always watching. Always listening. Always ready to respond to what’s going on in front of it.

Already, AI-powered virtual reality environments have allowed performers to act out scenes in immersive and realistic settings, expanding their range of possibilities and enhancing the audience’s experience. AI tech could also be used to generate real-time sound effects or music cues that could add depth and emotion to the scenes.

BUT… Can this soulless mechanization create anything of narrative value for putting on stage?

Here’s the story I asked AI- ChatGPT to create during my dinner in Cairo.  In less than 5 seconds it told me this story:

ME: Tell me a story about an improviser saving the world with improvisation story skills.

ChatGPT: Once upon a time, in a world not unlike our own, there was a man named Max who was an improviser, creating scenes and stories on the spot without any preparation. Max was known for his wit, humor, and ability to touch the hearts of his audience with his spontaneous storytelling.

One day, a great danger threatened the world. A giant asteroid was on a collision course with Earth and would cause massive destruction if it were to hit. The governments of the world were frantically trying to find a solution, but to no avail.

An idea struck Max. He believed that he could use the power of his storytelling to save the world. He approached the leaders of the world and offered to use his improvisational skills to create an emotional story that would inspire people to come together and find a way to stop the asteroid.

The leaders were skeptical at first, but Max was determined. 

He took to the stage in front of a global audience and began to weave his tale. He started with a story of a small community coming together to face a common threat and, through their love and determination, they were able to overcome impossible odds.

As he continued his story, people all over the world were moved by his words. They were inspired to join together and work towards a common goal. Scientists, engineers, and everyday people from all walks of life came together to find a solution.

In a remarkable feat of human collaboration, they developed a plan to deflect the asteroid using a massive thruster system powered by the world’s largest fusion reactors. It was a risky plan, but with the world united in their efforts, they were able to successfully launch the thruster system and change the asteroid’s trajectory, saving the world from certain destruction.

Max’s emotional story had truly brought the world together and saved it from certain doom. From that day on, he was hailed as a hero and his story was told and retold for generations to come. The power of storytelling and the human spirit to overcome adversity had been demonstrated for all to see.

Max may have been just an improviser, but in that moment, he used his unique talents to create a story that would change the world forever.

Sure, it’s simplistic, and doesn’t offer more than the flimsiest details but… like any improviser, it adapts. Bring it to a rehearsal and coach it to find more resistance and more details and you would have something worth watching.

I followed up the story with this request:
“Can you rewrite the story with more detail and more resistance to the hero’s success?”

it got a little better. The meteor was replaced with a villain. The villain had a name and a plan. For some reason it chose to rename the hero TOM, The story was still a little simple. I finally asked for one more attempt: 

“Can you make this story a tragedy where Tom seems to win but in a twist of fate is defeated?”

I won’t print the entire response but here’s the final paragraph in the revised story where the Villain seems defeated BUT was able to destroy everything, including Tom.

‘The world was plunged into darkness, and the only thing that remained was the memory of Tom’s bravery and his belief in the power of a good story. Tom had won the battle, but in the end, he had paid the ultimate price.”

It’s not great literature.  I think Nader’s profession is safe… for now. But AI does offer some interesting possibilities and even something more engaging than some of the Improvisation groups I know.

In Part 2 – I will tell you how AI is being used on stages around the world now, including some formats being explored by Improvisers like: Tim Tkachev, a theatre director & teacher based in Tbilisi, Georgia and co-founder of an international art Group 1212.

Part 3 – will offer a little glimpse into the future of the AI IMPROVISER including the “mistakes” that have already lead to real time character changes in AI that have threatened, and fallen in love with their “stage partners”


Consider that AI systems can be trained on human input, including language, stories, and other creative works, which can help them generate responses that are more natural and human-like.

While there is a common misconception that AI-generated content is soulless or lacks creativity, the truth is that AI systems are increasingly capable of producing novel and innovative content that can inspire and engage audiences.

Additionally, AI can be used in conjunction with human improvisers to create collaborative performances that are more dynamic and interactive. For example, an AI system can generate music or text prompts that human improvisers can use to build on their performances or suggest new directions for the story.

That last section was written entirely by AI.


If I’m being entirely honest with you… more than 70% of this article was co-created with two AI partners ChatGBT and RYTR. And 4 of the six pictures were created by AI based on what I asked for it to create.

In this article, can you guess what is real and what is not real and what is real and what is real and what is real and what is real and ….and….. and….  – error

What is real?






Leave a Reply

Avatar placeholder

Your email address will not be published. Required fields are marked *