It’s impossible to discuss the future of any practice or concept without discussing the impact of technology. While storytelling may be one of the most innately human acts there is, technology has played a role in the way we tell and preserve stories for a long time. Throughout history humans have used a variety of tools and technology to tell and preserve stories, from ochre on cave walls to images taken by cameras to documents stored computer databases.
The most significant impact technology will have on storytelling going into the future is the development of systems that cannot only record our stories but collaborate in their creation or create them entirely themselves. However, this is not just a concept for the distant future; machines already work with humans to tell stories, whether by design or coincidence.
The term ‘emergent storytelling’ is used to describe this relationship between human and machine. An emergent story is one not authored by a single person, but instead resulting from interactions between individuals and systems. Most commonly, emergent storytelling is used in the context of video games.
For example, Legend of Zelda: Breath of the Wild, released by Nintendo in 2017, has been hailed as a prime example of emergent gameplay. The abilities and tools the main character, Link, has access to can be combined in various ways by players to complete puzzles and quests in unique ways not even necessarily intended by the developers, allowing players to complete the story in the order and with the level of detail that they wish.
A more tangible example of an emergent storytelling device is the GPT-2, a language model created by artificial intelligence research laboratory Open AI.
GPT-2 generates coherent text from prompts given by users. The system was trained on over 8 million documents sourced from the internet and is available to the public through sites such as Talk to Transformer.
After typing in a phrase, the GPT-2 engine completes a following paragraph word by word, undertaking a form of emergent storytelling between artificial intelligence and human. While GPT-2 is a clear example of a system that can take part in emergent storytelling, users have identified something missing from the text it generates.
Following its release in 2019, programmer Janelle Shane used GPT-2 to create a project for National Novel Writing Month. She found that GPT-2 had limitations.
‘Some GPT-2 sentences were so well-crafted I wondered if they were plagiarised. Otherwise, the computer often journeyed into a realm of dull repetition or uncomprehending surrealism.’
It is clear that there is a human quality to storytelling and the way it evokes emotion that some argue can never be replicated by technology.
To ascertain whether technology will be developed that can create a genuine story on its own, we have to define what storytelling is. A stated in my first blog in this series, storytelling is ‘the social and cultural activity of sharing stories …. [that has] allowed history, religion, morals and cultural tradition to be passed on throughout the ages.’ Considering this, the question of whether artificial intelligence can write a genuine story becomes more philosophical.
Will allowing technology to create our stories for us dilute the importance of storytelling in our society and strip it of its benefits like recording history and teaching morals and values? Or will technology enhance our human ability to tell stories?
These questions are complex, but are ones we must ask ourselves as futurists as we move forward and storytelling continues to be an innate part of human nature and our society.