Embracing Imperfection: AI in Academia
“All opinions are not equal. Some are a very great deal more robust, sophisticated, and well supported in logic and argument than others.”
― Douglas Adams, The Salmon of Doubt
Let’s give credit when it is due. Avi Staiman, thank you for inspiring us to write this.
Avi (a Chef at Scholarly Kitchen and CEO of Academic Language Experts) opened a meticulously researched article on AI and academic publishing with a most memorable line:
“AI was most definitely used in writing this article.”
Avi’s article captures the essence of one of the hottest debates ongoing in the academic and research circles — should academia, and thus the field of education, fend off or embrace the use of AI? He elaborates as to why academic journals should not jump the gun with generative AI (Gen AI) content-detection tools considering the high false-positive rates with many (if not all) of them, including the one offered by Turnitin, the paragon of plagiarism detection.
Arguably the most riveting recommendation in the article is delivered in its climax. Avi rightly implores the publishing industry to utilize AI tools and technology to level the playing field for English-as-second-language scholars, thus promoting DEI, as well as aiding basic education processes. Both, of course, with the employment of AI guardrails, if one is wondering.
This is deeply refreshing. The academic community should technically lead society into embracing, even creating, change, and progress with the right checks and balances. Yet, it has been hesitant to take a measured stance to embrace AI the right way, while the rest of the world is welcoming AI in every imaginable aspect with reckless abandon. Both are equally worrying.
Aren’t we being naive to try to prevent future generations of students, who are likely to be even more tech-savvy than the existing ones, from using the best tools and resources available to learn and generate learning? Why resist technological change (for the better, and when done the right way) if it will make us more productive, efficient, and creative with things that we need to do?
What’s happening within academia?
Like Avi, we think that universities, scholarly societies, and other related institutions need to up the ante by drafting policies for conscious (dare we say educated) integration of AI into all academic and research activities. Don’t turn a blind eye toward AI integration. Even worse, don’t go guns blazing and ban any AI use as a breach of the code of conduct. Instead, let’s all sit down and acknowledge that we have all been using AI (maybe not generative) for the best part of three decades on most of our devices for most of the creative/generative work we have ever done.
In our analysis of university policies, we found a ubiquitous reluctance toward AI integration in the top 30 universities in the world with some notable exceptions. In the US, Harvard is leading the way in AI adoption by not only creating a centralized site for AI resources, and an AI institute but going as far as to create AI sandboxes for researchers to integrate AI into potentially sensitive research. Yale and the Russell Group universities (UK) are not far behind in developing holistic policies that do not suppress the use of AI.
We also happened to listen in on a conversation between Sam Altman, CEO of OpenAI, and Ben Nelson, Chairman and CEO of the Minerva Project, at World Knowledge Forum 2023. Conflicts of interest aside, Sam made a keen point that the educational system needs to adapt to the change more than Gen AI needs to adapt to educational needs. Universities and institutions, are you listening in?
Pushing the limits of education with AI
One of the predominant pushbacks from academia is that AI use may lead to a lack of original thought in academic writing and research. But this calls into question what originality is and how we assess it.
It must not come to anyone as a surprise that one’s ability to explore the unknown or to create the uncreated is trained and prompted by the knowledge we gain, the stories we read, and the things we watch and listen to. Assuming your jaw is un-dropped, ask yourself why AI, or specifically generative AI, is any different.
Why not consider AI as another tool that refines and expands on your line of questioning and reason and assists you in increasing the human generative potential? At the end of the day, we humans will still have the power to choose the best or the most logical outcome from the smorgasbord of options we may be presented with.
Moreover, such technology will level the playing field for students with poor writing or planning skills, which can be argued as ancillary and not primary skills for the advancement of knowledge. Why not then test everyone on their creativity and originality of thought with comparative assessments with/without these tools? If institutions provide opportunities for legitimate use of AI sandboxes, you could even set up generative AI tools and customized, one-on-one peer reviewers.
Alethea, the first of Clarivate’s acquisitions under its Academia & Government Innovation Incubator perfectly underscores this thought. Alethea is meant to facilitate meaningful engagement in a classroom setting through personalized and adaptive guidance on academic texts, assessments, and activities. It is also designed to assist in providing insights to faculty and librarians at an accelerated pace such that they can adapt curricula to the students’ needs.
All of this begs the question – why are educators holding back on AI integration, even for simple, straightforward tasks such as paraphrasing, editing, and literature review? Are they searching for the perfect tool that does not require due diligence or is it the very concept for augmenting learning with such an efficient one?
We’ll continue this conversation in our next post which focuses on the iterative nature of developing and deploying AI tools in academia.
Leave us your thoughts on ours and read on to Part 2!
