Beware of ChatGPT Trying to Teach Your Kids Math…or Anything

Wednesday marks a day of teacher strikes across much of the UK, putting parents in the familiar pandemic-inspired role of homeschoolers-in-chief for their children. Except this time, there’s a magical automated assistant at the ready to help you out.

Educators have cautiously praised ChatGPT, OpenAI’s ultra-sophisticated chatbot, saying it could revolutionize education. A headteacher in Britain says it sparked a homework overhaul, while another in Oregon used it to create lesson plans and study guides.

The tool’s personalized responses are what make it so tempting as an all-knowing digital tutor. I recently used it to dig deeper into the subject of enzymes, when my 12 year old son had some questions I had no hope of answering. When ChatGPT offered a dense and technical explanation, I asked him for simpler terms and an analogy.

“Safe!” he has answered. “Think of a lock on a door. The lock is like an enzyme and the key is like the substrate molecule…” He took the analogy further to describe the active site of an enzyme as the keyhole.

These were remarkable responses. We could have delved into all facets of biochemistry if we had wanted to. Unlike a human tutor, ChatGPT can be queried for as long as you want.

This holds enormous potential for personalized, independent learning…except that ChatGPT often gets it wrong, and it does a pretty good job of hiding it. When I tested one of my daughter’s English homework questions on the tool, it offered an eloquent list of examples, which on closer inspection included one that was extremely inaccurate. The main character had a turbulent relationship with his parents, the bot said, even though the character’s parents were dead throughout the book.

On another occasion, I used the tool to generate linear equations for my daughter to practice with. She was perplexed when I asked the tool to generate the answers, which were different from the ones she had calculated. I asked ChatGPT for an explanation and he once again broke down his method in simple terms, sounding as authoritative as any real math tutor. But when I double-checked the answers on Google, it turned out that ChatGPT’s answers were wrong and my tween’s were correct. Thus ended his mini math failure nightmare and much of my initial enthusiasm for ChatGPT.

New York’s public school system, the largest in the United States, has already banned its students from using ChatGPT, in part because of concerns about “content accuracy.” This is why recent comparisons of ChatGPT with a “calculator for writing” are a misleading analogy, since calculators are always right and ChatGPT is not.

How inaccurate is it? An OpenAI spokeswoman said the company had updated ChatGPT over the past two months to improve its factual accuracy, but had no statistics to share. The tool also warns users when they first open that it sometimes makes mistakes.

Will it be more precise? Yes, but it’s hard to say how much. The large language model that underpins ChatGPT is made up of 175 billion parameters, which are parameters used to make the model’s predictions, compared to the 1.5 billion its GPT-2 predecessor had. It has become accepted wisdom in AI that the more parameters you add to a model, the more truthful it becomes, and the correlation is real for GPT. It became much more accurate when all these parameters were added. The next iteration slated for release this year, called GPT-4, is said to have trillions.

The problem is, we don’t know if a huge jump in settings also means a huge jump in reliability. This is why students should use ChatGPT with caution, if at all, for the foreseeable future.

When I asked Julien Cornebise, Honorary Professor of Computing at University College London, if he would ever trust it as a homework tool, he replied: “Absolutely not, not yet. He pointed out that even when the system improves, we will still not have the guarantee that it is truthful.

Students should get used to corroborating any facts the system shares with other information online or with an expert. Albert Meige, associate director specializing in technology at consulting firm Arthur D. Little, says his own teenage daughter used him to help with her physics homework – but he was able to validate the answers thanks to his doctorate in computational physics. He recommends using the chatbot to help better understand questions in assignments. “She discovered that she shouldn’t ask a single question,” he says. “It was an interactive process.”

Use it to get feedback, Cornebise agrees. “That’s what the star student will do.”

Being a relatively small company, OpenAI can get away with spitting out the odd alternate fact. Google and Alphabet Inc.’s Meta Platforms Inc. couldn’t do the same. Google has its own highly sophisticated language model called LaMDA, but is extremely cautious about integrating a similar chatbot into its own search engine, likely in part because of the accuracy issue. Three days after publishing an AI tool capable of generating scientific papers, called Galactica, Meta pulled it after academics criticized it for generating unreliable information.

OpenAI will be held to equally high standards as the generative AI arms race intensifies and chatbot technology is integrated into search engines in the United States and China.

Until then, use it with discretion and a healthy dose of skepticism, especially in education.

© 2023 Bloomberg LP