In a recent article in The Conversation, Geoffrey Rockwell describes introducing his students to their new tutor — a chatbot.
The professor of philosophy and digital humanities wanted to resurrect a learning model as old as antiquity, so he created an ethics professor using Character.AI, a service that lets you create a fictional character you can engage in dialogue.
Rockwell asked his students to write a series of questions meant to generate a fruitful conversation on ethics with the chatbot — one that would itself raise questions about “what a dialogue is supposed to do and how it can be used to convey ideas” while encouraging a healthy skepticism about what the chatbot says in response.
After all, there is a long history of learning through dialogue that goes back at least as far as Plato and Xenophon, Rockwell points out, when Socrates was depicted conversing with his students.
“With the release of sophisticated chatbots like OpenAI’s ChatGPT … dialogue’s time has come around again,” writes Rockwell. “I suggest that we can make a virtue of the availability of these chattering machines.
“By thinking through dialogue, we could all rediscover the rich history and potential of this form of engagement. At the same time, we also have to teach our students to be careful and think critically about engaging with AIs and assessing the credibility of what they say.”
Geoffrey Rockwell and Alona Fyshe, a computing science professor and Canada CIFAR AI Chair at Amii, discuss Rockwell’s thoughts on dialogues with AI, computer ethics and the need for more AI literacy.
Rockwell has spent decades interrogating how a rapidly evolving digital culture has changed what it means to be human. He is a leader in the field of digital humanities, integrating computational methods and theories with research and teaching in the liberal arts.
Now, Rockwell has been appointed as a Canada CIFAR AI Chair at Amii, a non-profit institute that supports world-leading research in AI and machine learning and translates scientific advancement into industry adoption.
The CIFAR program was launched in 2017 to bolster the federal government’s Pan-Canadian Artificial Intelligence Strategy to “drive the adoption of artificial intelligence across Canada’s economy and society” and recruit and retain AI talent in Canada.
More than 120 chairs have so far been appointed at Canada’s three national AI institutes, including Amii, Mila in Montreal and the Vector Institute in Toronto. The U of A’s Martha White of the Faculty of Science also saw her chair at Amii renewed.
Rockwell says his appointment reflects CIFAR’s attempt to widen its scope beyond computer science.
“For the first couple of years, the chairs were in computer science, but now there’s this idea of expanding to everything from ethics to biology to Indigeneity,” he says.
Rockwell’s research interests include video games, textual visualization, text analysis, ethics of technology and ethics of artificial intelligence. He is now working on a book on dialogue and artificial intelligence, exploring the cultural discourse of AI as well as imagining conversations we might have with new chatbots such as ChatGPT.
He adds that chatbots are in some ways just the latest phase in our historical fascination with dialogue.
“One of the things that struck me about ChatGPT is that nobody was paying a lot of attention to AI until a large language model was developed and optimized for dialogue. That’s when we saw this explosive interest.”
In 2022 he won the international Antonio Zampolli Prize with Stéfan Sinclair for co-developing a suite of text analysis and visualization tools called Voyant Tools, and was project leader for TAPoR, a text analysis portal for research funded by the Canada Foundation for Innovation.
He published Defining Dialogue: From Socrates to the Internet (2003) and Hermeneutica (2022), also with Sinclair, and is co-author of Right Research: Modelling Sustainable Research Practices in the Anthropocene (2021).
“As the science and application of AI continues to advance rapidly, the ability to attract and retain great researchers who are advancing diverse aspects of the field will be critical to Canada’s leadership in developing and deploying responsible AI,” says Elissa Strome, executive director of the Pan-Canadian AI Strategy at CIFAR, of the new AI chairs.