AI Tutors: Can Chatbots Revolutionize Education?

⏱️ 13 min read                  

The future school classroom is about to go through a transition. Every student has a personal tutor—available 24/7, customizing lessons just for them. Sounds like a dream, right? Well, Silicon Valley believes AI chatbots could make that dream a reality. But can these tools really live up to the hype, or are we chasing another hi-tech fantasy? There’s promise and pitfalls in the world of AI-powered education. Let’s explore this, shall we. 

Welcome to That's Life, I Swear. This podcast is about life's happenings in this world that conjure up such words as intriguing, frightening, life-changing, inspiring, and more. I'm Rick Barron, your host. 

That said, here's the rest of this story 

During a compelling presentation at TED, Khan Academy founder and CEO Sal Khan shared his vision for an AI-powered revolution in learning.

Khan Academy founder and CEO Sal Khan, predictd last year that A.I. tutoring bots would soon revolutionize education. Big Tech executives have similar visions. Courtesy of New York Times

"We're standing at the threshold of what could be education's most transformative moment yet," proclaimed Khan, whose digital learning platform has educated countless students worldwide. His bold vision? "To provide every student across the globe with their own personal AI tutor – one that's both artificial and extraordinary."

The online community responded enthusiastically to Khan's presentation about AI tutoring, with view counts reaching into the millions. This sparked a chain reaction among technology leaders, with Google CEO Sundar Pichai among those echoing similar sentiments about AI's educational potential.

Following Khan's presentation, Pichai appeared on Harvard Business Review's podcast, stating, "Looking ahead, I envision a future where AI tutoring becomes universally accessible – reaching every individual, regardless of their background or location." This aligned with Google's ventures into AI, including their chatbot Bard, launched in 2023. The tech giant had also demonstrated its commitment to digital education through substantial financial support to Khan Academy, exceeding $10 million.

The concept of AI-powered tutoring that Khan champions reflects a long-standing aspiration within the technology sector: creating automated educational systems that adapt instruction to each student's unique needs. Supporters believe these adaptive learning platforms could bridge educational disparities by delivering personalized content to students faster and more efficiently than traditional teaching methods.

This technological pursuit of educational excellence has manifested in various initiatives over time. Industry leaders and charitable foundations have promoted one-to-one device programs in schools, supported digital learning platforms, and invested in adaptive learning applications. While certain digital programs targeting mathematics and reading skills have shown promising outcomes, the broader impact of educational technology has often fallen short of expectations. Many of these digital learning initiatives haven't substantially succeeded in reducing academic performance gaps or improving key metrics such as high school completion rates.

The emergence of advanced AI systems that generate content - from scientific explanations to written analyses - has sparked fresh optimism about automated education platforms. However, skeptics point out that concrete evidence supporting the educational benefits of AI tutoring systems remains to be proven.

Students can use a new A.I. tutoring bot developed by Khan Academy. Courtesy of New York Times

Leading digital education providers have begun integrating chatbot tutors powered by OpenAI's GPT-4 technology into their platforms. Both Khan Academy and Duolingo are pioneering this approach, utilizing the sophisticated language model that processes vast text databases to generate contextually appropriate responses to user inputs.

Technology industry leaders envision these AI systems eventually developing the capacity to connect with and motivate students in ways that mirror effective human educators. OpenAI's president Greg Brockman expressed this sentiment during an appearance on the "Possible" podcast (co-hosted by OpenAI early investor Reid Hoffman) in the previous summer: "Consider the implications of providing every student unlimited access to this caliber of teaching, completely free of charge," he mused. While acknowledging the ambitious nature of this goal, Brockman noted that "what once seemed purely theoretical is increasingly within our grasp."

The Biden administration embraced AI's educational potential, as evidenced by an AI executive directive. According to official White House documentation, the administration sought to leverage AI's capabilities in education through initiatives, supporting teachers' use of 
AI-enhanced learning tools, particularly in individualized tutoring.

However, education specialists urge caution regarding the widespread adoption of AI in classrooms. Their concerns stem from several key issues inherent to current AI technology.

Academic experts highlight significant limitations of AI chatbots:

·       including their tendency to generate inaccurate information without hesitation. 

·       There's growing concern that incorporating these AI systems as standard educational tools could legitimize potentially unreliable sources of information in academic settings.

·       Additionally, critics emphasize two other crucial problems: the potential for bias within AI systems and their "black box" nature, making it difficult for educators and students to understand how these tools arrive at their responses.

According to Ben Williamson, who serves as a chancellor's fellow at the University of Edinburgh's Centre for Research in Digital Education, integrating generative AI tools in education might impede rather than enhance student learning outcomes.

Ben states, "We're witnessing an eagerness to embrace these AI chatbots and their underlying technology as trustworthy educational tools," Williamson notes, "yet we lack concrete evidence supporting their effectiveness in achieving claimed educational benefits."

Education experts raise another significant point: the intense focus on experimental AI tutoring systems could divert attention and resources from proven educational interventions. Programs like universal preschool access have successfully boosted graduation rates and college enrollment.

The deployment of AI in education also raises complex ethical and legal considerations. A key issue involves the training methods of large language models, which utilize internet-sourced texts without providing compensation to content creators. This practice has prompted legal action, including a lawsuit by The New York Times against OpenAI and Microsoft. This approach particularly concerns teachers' unions advocating for fair compensation.

Additional apprehensions center on data usage practices. There's growing unease about how AI companies might exploit educational interactions - including teacher inputs and student responses - to enhance their commercial AI systems.

The American Federation of Teachers, led by Randi Weingarten and representing over 1.7 million educators, actively collaborates with congressional leaders to develop regulatory frameworks for educational AI technologies.

Randi states, "Our members routinely integrate educational technology into their teaching," Weingarten explains. "They're seeking greater input into how these tools are implemented in their classrooms. We're striving to maximize AI's benefits while implementing safeguards against potential risks."

The current enthusiasm for automated teaching solutions echoes similar movements from past decades. During the 1960s, educators and innovators applauded the arrival of "teaching machines" - mechanical and electronic devices programmed to deliver individualized instruction in subjects like mathematics and spelling. These were predicted to transform education fundamentally.

This technological optimism was captured in an October 1961 Popular Mechanics article titled "Will Robots Teach Your Children?" The piece described a nationwide surge in machine-based learning experiments, where students worked independently with devices that allowed them to progress at their pace. The article also highlighted concerns that the new machines posed significant challenges for teachers and students. It questioned whether educators would merely become "overqualified babysitters" and asked, “How might machine-based learning affect students’ ability to think critically?”

Though clunky and overly formal, these early teaching devices were a fleeting trend in classrooms, overhyped by some and feared by others. A similar story is unfolding with the introduction of A.I. teaching bots, which are viewed as both transformative and potentially harmful to education.

However, unlike the rigid teaching machines of the 20th century, A.I. chatbots are more flexible and adaptive. They generate real-time, conversational responses tailored to individual students, making learning more engaging and enjoyable.

Some advocates see these A.I. tutoring bots as helpful companions for students, offering discreet assistance without fearing embarrassment. Should schools widely adopt such technology, it could significantly reshape how children learn.

This potential has led some former tech industry leaders to shift their focus to education. Jerome Pesenti, formerly Meta’s vice president of A.I., has launched Sizzle A.I., a tutoring service. The app’s chatbot uses a multiple-choice system to guide students through math and science problems.

Jared Grusd, formerly Snap's chief strategy officer, has co-founded Ethiqly to assist students with their writing. The app's A.I. chatbot is designed to help students outline and structure their essays while providing feedback on their drafts.

One of the most prominent advocates for tutoring bots is Sal Khan. His educational platform, Khan Academy, introduced an A.I. tool called Khanmigo in 2023, specifically tailored for classroom settings. Rather than completing student assignments, Khanmigo guides them through problem-solving processes in subjects like math.

Students at Khan Lab School were among the first to try Khanmigo, a new A.I. tutoring bot that can help students work through math problems and other subjects. Courtesy of New York Times

Khanmigo also keeps records of the interactions between students and the bot, allowing teachers to review them later. A clear disclaimer on the platform reminds users: "Khanmigo makes mistakes sometimes." Schools in Indiana, New Jersey, and several other states are testing this chatbot tutor as part of pilot programs.

Sal Khan’s inspiration for tutoring bots partly comes from science fiction, specifically Neal Stephenson’s cyberpunk novel The Diamond Age. In the story, a futuristic tablet-like device can teach an orphaned girl precisely what she needs at the right time by analyzing her voice, facial expressions, and environment in real-time.

Khan believes that within the next five years, tutoring bots like Khanmigo could offer a similar level of personalization, equipped with privacy and security measures.

“The A.I. will be able to observe a student’s expression and say, ‘You seem a bit distracted. Let’s refocus,’” Khan explained, envisioning how these bots might soon interact with students. 

What can we learn from this story? What's the takeaway?

Sal Khan, founder and CEO of Khan Academy, believes AI could revolutionize education. He highlights the potential for AI to provide every student with a personal tutor and every teacher with an assistant. 

Khan said he hopes to harness the potential influence of AI usage to direct attention toward positive educational growth, pointing to efforts by social media corporations and search engines to compete for consumer attention using AI.

Well, there you go, my friends; that's life, I swear

For further information regarding the material covered in this episode, I invite you to visit my website, which you can find on Apple Podcasts/iTunes for show notes and the episode transcript.

As always, I thank you for the privilege of you listening and your interest. 

Be sure to subscribe here or wherever you get your podcast so you don't miss an episode. 

See you soon.