Don't Ban AI in Education, Introduce AI Literacy instead
Image generated via Canva's text-to-image tool

Don't Ban AI in Education, Introduce AI Literacy instead

Disclaimer: All opinions are my own and do not reflect my employer’s. My experience is limited to GPT-3 and not GPT-4


When ChatGPT was unleashed into the world, I found myself wondering how it will change the future of education. Reflecting back on my own journey, I thought about how after 3 tertiary-level degrees, what was truly essential to my establishing and growing my career was the perennial skill of learning how to learn.

I’ve always argued that schools don’t do a great job at teaching us the real skills we need to navigate life. Instead, we gather so much specialized knowledge about a single field, and if one day we find ourselves in a country where that knowledge is practically irrelevant, then we’re pretty much on our own.

But then again, I’m very, very biased here.

So enter Artificial Intelligence (AI), and all of a sudden, students are flocking to chatGPT asking it to generate essays and do their homework. It’s ruffling feathers in academic institutions with some of them completely banning it (Link), and others using software to catch students whose essays sound like they’re AI-generated, even if they aren’t (Link).

As a person who’s been actively using chatGPT to speed up my own learning, I’m of the opinion that rather than blanket bans of AI, academic institutions should teach with it. Banning generative technologies is akin to forcing people to ride horses in a world where cars not only exist, but are also free. It will only widen the gap between people who use this technology early on and those who don’t.

Let's just say that the latter group won't reach as far or as fast as the former group.

What we need more of are not blanket bans but rather AI literacy.

Just like principles in internet literacy came along with the advent of the internet, principles in AI literacy need to be more commonplace. I remember a time when the proliferation of fake news was rampant, and this infographic started spreading teaching people how to spot fake news by considering the source, checking the author and date of the article, and also searching for supporting sources.

This is easier to do when you have a list of articles, and you’re able to see the individual sources, since you’re able to mentally give different weights to each source - or dismiss them completely.

However, the model has changed now with ChatGPT where you only get one answer. And lo and behold, that answer appears in elegant prose. Even if it doesn’t make sense.

Take for example this one time when I was using it to check my understanding on a particular concept. It confidently - and fluently - went against the transitive law by telling me how A = B, and C = B. So when I asked it if A and C were the same?

It said, “no”.

I tried to reason with it, and it held onto its stance very stubbornly, insisting that A ≠ C even though two paragraphs before, it showed me how both were equal to B.

So that’s when I realized the technology still has some way to go.

Back to AI literacy.

I just asked chatGPT to give me some principles for AI literacy, and this is a snippet of its answer:

  1. Recognizing the potential benefits and risks of AI, including issues related to bias, transparency, and accountability.
  2. Understanding the ethical considerations of AI development and use, including issues related to privacy, surveillance, and human rights.
  3. Being able to evaluate AI systems critically, including their performance, reliability, and potential impact on human decision-making.

I want to end this article by stating that I am optimistic with regards to AI’s impact on education. For the longest time, we’ve been stuck in a system built on memorizing facts and regurgitated them onto exam papers for grades. Maybe now, we’ll finally focus more on higher-level cognitive skills such as critical thinking, creativity, and problem-solving…once we reach a certain level of accuracy with such technologies that doesn’t require us to fact-check everything. It could also move us away from the silo-ed specialized nature of education and back to the interdisciplinary one, where we use knowledge from multiple fields in order to solve real problems and add value to the world.

What are your thoughts?

Interested in joining my mailing list??Add your email address here: https://eepurl.com/gn_RcD. I don't send weekly newsletters but do send other special things every once in a while.

Nyambura Gachahi

Professional Virtual/Executive Assistant | Junior Climate Scientist | Data Scientist | ALX Fellowship Ambassador | Social Media Manager/analyst | Social Justice Agitator | Opportunity Connector/getter

1 年

That you are absolutely right. AI is like how you learn to write from drawing gibberish to connecting letters, words and finally full sentences We absolutely need to exist in a world where we interact with it as the tool and human creation it is, and I reiterated the very same during today's event from The Room ???? ??We need to also find ways to teach children and older people to interact with it as well as they learn the ways it can also be abused and weaponised against them. I personally fully believe in knowledge but filtered for all ages and stages of education and life???

回复

要查看或添加评论,请登录

社区洞察

其他会员也浏览了