Clarification: This story has been clarified to reflect the fact that ISTE is collaborating with Google on the development of StretchAI. ISTE has also worked with other organizations, including OpenAI, on prototypes of future tools for supporting educators.
ChatGPT can spit out a coherent sounding paper on the causes of the Civil War or analyze a Shakespeare play in seconds. But it may flub the most basic facts about Abraham Lincoln or Hamlet when generating those essays.
The artificial intelligence tool pulls from nearly every imaginable source on the internet, even if much of the content is not accurate or produced by a reputable source. That鈥檚 not exactly the kind of technology that will help earn educators鈥� trust.
Enter a new, more focused version of the technology that some call 鈥渨alled garden鈥� AI and its close cousin, carefully engineered chatbots.
These cutting-edge bots are similar to generative large language models like ChatGPT in that they are trained using internet data. But instead of absorbing large swaths of the internet and treating it all somewhat similarly, they generate feedback based on a more limited database of information that their creators deem reliable.
Put another way: If ChatGPT and other more general large language models are the vast and largely lawless Wild, Wild West, this walled garden breed is a small, one-room K-12 school house run by a very strict and discerning teacher.
鈥淧eople understand that a general model isn鈥檛 going to serve the needs of education as well as specialized models,鈥� said Joseph South, the chief innovation officer at the International Society for Technology in Education, a nonprofit, and one of the first education organizations to experiment with walled garden AI and specialized chatbots. 鈥淲hat makes them special and different is that the content that goes into them is curated. So, you don鈥檛 face all the dark side of the internet in your model that you have to filter out. You never put it in.鈥�
If people are promising you a perfectly safe chatbot, you should probably be skeptical for the time being. But it's exciting. This is breaking some new ground.
ISTE, which recently merged with ASCD, is working to develop Stretch, a chatbot trained only on information that was created or blessed by the two professional development organizations. This walled-garden model鈥攚hich is not yet available to the general public鈥攃an also cite its sources, giving users鈥� a digital trail to follow in gauging its accuracy, according to South.
That ability, along with the more curated training data, means Stretch and similar bots could be 鈥渋ncredibly useful, incredibly aligned [to educators鈥� needs] and much safer to deploy鈥� than ChatGPT and its ilk, South said.
Stretch, which is being developed in partnership with Google, is an early example of the kind of chatbot that teachers and students are likely to become very familiar with, South predicts.
These more focused bots will likely help teachers find lesson plans tailored to students鈥� needs and interests, tap research-based professional development, and even serve as a kind of high-tech tutor for students, South said. Similar technology is already available or will be developed for fields like healthcare and energy, he added.
While nonprofits like ISTE will have their own bots, so will other organizations, South predicted.
鈥淭here鈥檚 going to be different gatekeepers for these models,鈥� South said. 鈥淎nd it鈥檚 going to matter a whole lot: who built the model? Because whoever built the model is providing you your universe of content that you will be drawing from.鈥�
That鈥檚 why Pam Amendola, an English teacher at Dawson County High School in Dawsonville, Ga., feels conflicted.
On the one hand, a bot focused on professional development from ISTE holds a lot promise and credibility, said Amendola, who completed ISTE鈥檚 special AI training program several years ago.
On the other hand, sorting through the biases and viewpoints of the developers who designed the algorithms that choose the data that a 鈥榳alled garden鈥� bot concentrates on opens up a whole new set of questions in the burgeoning field of AI literacy, she added.
鈥淲e should be questioning everything鈥� and not assuming bots are all-knowing, she said.
鈥楤reaking new ground鈥�
This kind of specialized AI may have exciting applications in schools, but it鈥檚 not going to be easy to create or to monitor, said Michael Littman, the division director of information and intelligent systems at the National Science Foundation and a professor of computer science at Brown University.
鈥淚 don鈥檛 know if I鈥檇 call it the future of AI but it could very well be the future of chatbots,鈥� Littman said. 鈥淭he biggest challenge at the moment is combining the fluency鈥� of more general models like ChatGPT 鈥渨ith the ability to actually talk about something particular,鈥� he explained.
But he added that just as 鈥渂ugs can fly into a [real-life] walled garden,鈥� these models will still have to work to keep out bad information or 鈥渉allucinations,鈥� computer-science speak for inaccuracies, Littman added.
鈥淚f people are promising you a perfectly safe chatbot, you should probably be skeptical for the time being,鈥� Littman said. 鈥淏ut it鈥檚 exciting. This is breaking some new ground.鈥�
South sees several different possibilities for schools with this more focused, and its creators hope, more reliable, AI. Teachers could use it to plan lessons鈥攆or instance, asking for ideas to help students with dyslexia learn how to summarize what they read, South said. That鈥檚 something teachers already do with more general large language models, but the focused bots would likely be more efficient and trustworthy.
Another possibility: A teacher may be seeking help with some aspect of their practice鈥攍earner differentiation, student wellness, classroom management鈥攂ut only want to consider evidence-backed information.
And walled-garden AI could control for an obvious problem: When it comes to teaching and learning, a strategy that鈥檚 all the rage one minute鈥攕ay, personalized learning鈥�might be debunked down the line. That鈥檚 why ISTE and ASCD are using just a short time window鈥檚 worth of data鈥攁round the two most recent years鈥攖o train Stretch, South said.
AI as the ultimate tutor?
Maybe the most powerful potential use of curated AI in schools: Tutoring. A good teacher helps illuminate information and concepts for students but they 鈥渃an鈥檛 anticipate the needs of every student who comes to their classroom,鈥� South said.
A carefully engineered, focused chatbot, however, could 鈥渆ngage students from where they are very specifically, and that can be incredibly powerful,鈥� South said. 鈥淎 teacher doesn鈥檛 have time to do that with every student. But AI does.鈥�
One of the most prominent early examples of this is Khanmigo, a chatbot developed by the nonprofit Khan Academy.
Khanmigo works differently from ISTE鈥檚 Stretch. Instead of training the tech to focus its information absorption on a carefully curated corner of the internet, Khan Academy has engineered its bot to act like a tutor.
Khanmigo doesn鈥檛 give students a direct answer to their questions. For instance, if a student is learning about long division and asked the bot to calculate, say, 10,864 divided by 342, the student wouldn鈥檛 get an immediate answer, Kristen DiCerbo, the chief learning officer at Khan Academy, said in an interview.
Instead, the bot might respond with something like: 鈥溾�榯hat鈥檚 a great question. Thank you for asking. Do you have any ideas about how to start thinking about getting an answer?鈥欌�� And the student might say 鈥溾�榳ell, I know I am supposed to look at the tens and the ones.鈥欌�� The bot will then say 鈥溾�榯hat鈥檚 a great way to think about it. Let鈥檚 talk about how to do that.鈥欌��
Or if the student says initially that they aren鈥檛 sure where to begin, the bot might point them in the right direction, again, without spitting out an answer.
Khanmigo can also personalize its instruction, giving students long division examples based on their interests, such as baseball or high fashion.
The tool is still in a testing mode, DiCerbo explained, and allows users to report instances where it doesn鈥檛 act like it鈥檚 supposed to. So far, only about 2 percent of its interactions have been flagged as problematic, DiCerbo said.
鈥淩ight now, we鈥檙e finding [a] large language model with guardrails around it is pretty successful,鈥� DiCerbo said. 鈥淲e鈥檒l see if we reach a limit on what we can do with it, but we haven鈥檛 so far.鈥�
