Advocates for Ethical Use and Guardrails

Sal Khan, the founder of the online learning platform Khan Academy, shared his insights on the Doha Debates podcast about the importance of implementing ethical guidelines and safeguards when integrating Generative AI into educational tools for students. Khan voiced concerns about the potential misuse of AI technology by profit-driven edtech (Educational Technology) companies.

In the episode, Khan engaged in a discussion with NBC News technology correspondent Jacob Ward, under the moderation of journalist Joshua Johnson. Khan, who transitioned from a hedge fund analyst to an educational innovator, established Khan Academy in 2008, inspired by his experience tutoring his cousins. His non-profit organization, which focuses on edtech, now reaches over 160 million users across 50 languages and has created more than 6,500 video lessons. The organization has recently introduced Khanmigo, an AI-based teaching assistant, currently being trialed with students in California.

Khan reflected on his early exposure to AI technology. “When OpenAI approached us over a year ago, before the emergence of ChatGPT, we were among the first to witness GPT-4,” revealed Khan. “It was astonishing to see its capability to deliver the kind of in-depth tutoring we’ve always aimed for, exceeding what I believed was achievable in my lifetime.”

Addressing the issue of academic dishonesty in AI, Khan lamented certain edtech startups potentially exploiting AI for unethical purposes. “Prior to the advent of Generative AI, there were already multi-billion dollar companies whose business model revolved around completing homework for students.”

Khan warned of the influx of investment into AI-driven edtech, cautioning that market pressures might eventually lead these companies to offer services that encourage academic dishonesty, rather than focusing on sound pedagogical practices.

He highlighted the unique challenges in sectors like education and healthcare, where market dynamics can lead to less than ideal outcomes due to the disconnect between beneficiaries, payers, and decision-makers.

Khan also addressed potential perils with AI in online education, including biases and the tendency of AI to generate incorrect information. He remarked, “It’s surprising to some that AI can be notoriously poor at mathematics.”

For college students considering AI tools, Khan advised caution, especially in contexts where their use might breach ethical standards or honor codes.

“These AI models are already performing impressively, scoring in the top percentiles of exams like the LSAT and surpassing many doctors in medical board exams,” emphasized Khan on the rapid changes in AI technology. “As we progress to GPT-6 or GPT-7, with complexities akin to the human brain, AI will pose new and significant questions.”

Khan stressed the importance of implementing safeguards in AI applications, as demonstrated by Khan Academy’s approach with Khanmigo, focusing on content grounded in human-verified information to mitigate misinformation risks.

In the educational setting, Khan reaffirmed Khan Academy’s commitment to using technology to foster more human interaction. “Our guiding principle is to use technology thoughtfully to enhance human engagement. The focus should be on active learning, not passive lecturing, which technology can facilitate more effectively.”

Khan also shared insights on the early stages of AI efficacy studies at Khan Academy, indicating promising signs of increased student engagement and perseverance.

He warned of the risks of neglecting the responsible use of AI, highlighting the potential for a widening educational divide and the importance of acting as a positive force in harnessing AI technology for the benefit of all students.

Jacob Ward, author of “The Loop: How Technology Is Creating a World Without Choices and How to Fight Back,” also offered a cautionary perspective. “Inherently, technology acts as a shortcut, often repurposing others’ work inaccurately or misleadingly, and operates as a ‘black box’ without transparency. This, coupled with market forces, can detrimentally impact education by reducing available resources and increasing productivity expectations.”

“Our goal is to be that good actor. Let’s harness these technologies to try to level that playing field and up the rigor for everyone,” concluded Khan. “There are so many people with such a strong incentive to create crap that is not going to be good for people, it’s going to create a lot of noise.”

The post Advocates for Ethical Use and Guardrails appeared first on AIPressRoom.