Let’s Look At This Clearly: AI Hype in Ed Tech

Change Management, EdTech, Generative AI, Speculative Fiction

I’ve had a front-row seat to the explosion of hype around artificial intelligence over the past year. While AI tools like ChatGPT and DALL-E can certainly be useful in specific circumstances, the breathless speculation and grandiose claims about AI’s capabilities are creating more confusion than clarity.

The Problem of AI Hype

Overpromising and Underdelivering

Many critics have pointed out how AI companies and boosters consistently overpromise and underdeliver. Ed Zitron, in a recent piece, notes how AI executives like OpenAI’s Sam Altman habitually speak in vague generalities about future capabilities without providing specifics:

“Every time Sam Altman speaks he almost immediately veers into the world of fan fiction, talking about both the general things that ‘AI’ could do and non-specifically where ChatGPT might or might not fit into that without ever describing a real-world use case.”

This pattern of hype without substance extends to many AI product launches. In a Wall Street Journal interview, OpenAI CTO Mira Murati struggled to provide basic details about their new Sora video generation tool, repeatedly falling back on vague promises that various capabilities would be available “eventually.”

The Media’s Role

The problem is compounded by media coverage that often fails to critically examine these claims. As Zitron puts it:

“The media has been fooled, in the same way they were fooled by the metaverse, by the specious promises of AI and the executives that champion it. The half-truths and magical thinking have spread far faster due to the fact that AI actually exists, and it’s much easier to imagine how it might change our lives, even if the way it might do so is somewhere between improbable and impossible.”

Impact on Education

Misunderstandings and Disappointment

In my work supporting districts with technology integration, I’ve seen firsthand how this hype cycle leads to misunderstandings about what AI tools can actually do. Educators often get excited about the possibility of AI revolutionizing education, only to be disappointed when they discover the reality is far more limited.

Key issues I’ve observed include:

Overestimating AI’s ability to personalize learning and the value of doing so
Unrealistic expectations about AI’s capacity to automate grading and feedback
Confusion about the role of AI in content creation and plagiarism detection

The Reality of AI Capabilities

It’s crucial to recognize that current AI models, despite impressive language abilities, do not actually reason or understand in any meaningful sense. Researchers Michael Townsen Hicks, James Humphries, and Joe Slater argue in a recent paper:

“LLMs are doing considerably less than what human brains do, and it is not clear whether they do what they do in the same way we do… Their primary goal, insofar as they have one, is to produce human-like text.”

They characterize the output of large language models as a form of “b*******” – text produced without any real concern for truth or accuracy. While this may seem harsh, it highlights how these models fundamentally differ from human cognition and reasoning.

Finding a Balanced Approach

Practical Applications in Education

Despite the limitations, AI tools can be valuable in educational settings when used appropriately. In my presentations to educators around the Midwest, I often discuss ways AI can augment and assist various educational tasks. Some promising applications include:

Early content curation, resource development, and formatting
Generating additional ideas for discussion prompts, activities, and more
Providing basic writing assistance for students
Answering general questions about a document, article, or knowledge base
Acting as a simple coach on a topic and roleplaying

However, I always emphasize the importance of understanding the limitations and potential pitfalls of these tools.

Developing AI Literacy

The incessant hype and speculation ultimately harms AI literacy among educators and students. When people are constantly told that superhuman AI is just around the corner, they lose sight of the reality of what current tools can actually do. This makes it harder to have grounded, productive conversations about the appropriate use of AI in education.

To combat this, we need to focus on developing true AI literacy in our schools. This involves:

Teaching critical evaluation of AI-generated content
Understanding the basic principles of how AI models work
Exploring the ethical implications of AI in education
Practicing responsible and effective use of AI tools

The Role of Science Fiction in Shaping AI Perceptions

At ISTE Live 2024, I presented on how science fiction influences our perceptions of generative AI, a topic crucial for understanding the current AI hype cycle. Science fiction has long shaped our expectations of future technologies, including AI. In my presentation, I highlighted three common AI depictions in science fiction: the big evil AI (like HAL from “2001: A Space Odyssey”), the flawless AI assistant (like Jarvis from “Iron Man”), and human-like AI (as seen in the movie “Her”).

These portrayals, while entertaining, can create unrealistic expectations about AI’s capabilities and potential impacts. As I noted in my talk, “AI, the extremes of AI are portrayed for clickbait or sensationalism and that’s because it’s exciting… The problem is it’s exciting and so it gets rolled right into the hype cycle around AI where we have this peak of inflated expectations.”

A key takeaway from my presentation was the need for critical thinking when it comes to AI: “At the end of the day we need to be viewing things and thinking about AI critically, not thinking in terms of pop culture or science fiction and instead trying to really think critically about what it is that we’re looking at.” This critical approach is essential for educators as they navigate the integration of AI tools in their classrooms and curricula.

I concluded by calling for more nuanced and positive depictions of AI in media that can help us envision futures where technology enhances education rather than replaces educators. This perspective aligns with our goal as educational technologists to find practical, ethical, and beneficial applications of AI in education.

Conclusion

To wrap up: as Ed Zitron notes, “Every single time I’ve read about the ‘amazing’ things that artificial intelligence can do, I see somebody attempting to add fuel to a fire that’s close to going out.” Perhaps it’s time we let that fire die down and focus on the practical realities of AI rather than fanciful visions of the future. In education and beyond, we need to cut through the noise and hype to develop true AI literacy.