Almost immediately after ChatGPT was first released in late 2022, alarm bells sounded in the education world about how it would negatively impact learning. That year, The Atlantic ran stories with headlines like “The College Essay Is Dead” and “The End of High School English.” The predictions were that unleashing AI to the world at large would lead to an exponential rise in cheating at school and that student essays would cease to be original.
“I mean since day one the narrative in education has been really focused on cheating. And AI is still mostly seen as that thing that kids are using to cheat. So, you know, let’s not be naive. There are lots of kids using AI to cheat,” says Leon Furze, an educational consultant and the author of Practical AI Strategies: Engaging with Generative AI in Education. But he adds, “We’ve got to kind of go beyond that, I think and meet the students where they’re at with the technology.”
Cut to 2025 and that is what’s happening at many of Los Angeles’ top private schools, where panic has been replaced with pragmatism.
Yes, there are guardrails. Many schools carve out AI-free zones for test-taking, essay writing and other creative endeavors. “A lot more work, especially writing assignments, is now done exclusively in class,” says showrunner and writer Anthony Sparks (Bel-Air, Queen Sugar), who has a child at the prestigious Harvard-Westlake School.
The same applies at L.A.’s Sierra Canyon School (where recent graduates have included Bronny and Bryce James). Says Noah B. Salomon, the school’s English department chair, “For most assignments, in order to preserve training in independent critical thinking, we create an environment where students cannot use AI [such as] in-class writing, participating, graded discussions.”
But as generative AI continues to seep into seemingly every facet of society, administrators and teachers are increasingly looking at creative ways to engage with the technology with an eye toward preparing students for the real world.
Last July, The Buckley School (which Laura Dern, Bret Easton Ellis and Paris Hilton once attended) held a one-week AI Intensive summer camp on campus ($1,500 for the week), where middle-school-age students could learn everything from “AI + Art” to how to “leverage machine learning to help first responders allocate resources in crisis situations.”
At Windward School (which Josh Groban and Zoe Kazan attended), a special education and tech team has been working on helping teachers deepen their understanding of ChatGPT, Google Gemini and MagicSchool AI (an education focused AI platform).
And at Sierra Canyon School, the administration has formed a partnership with AI education platform edYOU, which provides students with personalized chatbot avatars that can help them with everything from tutoring to tracking their coursework.
Sierra Canyon’s Salomon details additional ways that the school is engaging students in the use of AI. “We have asked teachers to experiment, to find ways to leverage the power of AI in their teaching and to encourage students to use AI in a way that is ethical and that does not interfere with their learning or their curiosity. For example, students in a literature class might be asked to have a conversation with ChatGPT about a particular chapter of a novel, but they will be required to use specific quotes from the novel to challenge ChatGPT’s initial reading. Students might be encouraged to use AI to create quizzes on material or to have AI review an essay to offer another source of feedback on structure.”
And for every student who wants to use AI as a crutch or shortcut, there are other students who are concerned about using it responsibly, even sparingly. “I had a 13-year-old student who asked me, ‘Even if my school teaches us how to use AI, should we?’” shares Furze. “And she was really concerned that if she starts using AI at 13 years old, that by the time she’s 18, she won’t be able to use her own brain anymore. There are really young students who are genuinely concerned about this. On the flip side, I’m also seeing that there’s a lot of pressure with students to use AI because they think they’ll be outcompeted. There’s a lot of perception that, ‘if I don’t use AI, the student next to me is going to use AI and they’re going to do better. So I have to use it to stay competitive.’”
Adds Stacie Muñoz, director of educational innovation and technology services at Oakwood School (whose alumni include Chris Pine and Lily-Rose Depp), “I think there is an assumption that all kids are using AI all the time and I will say that I have talked to many kids who are not falling in that camp and are really hesitant about AI and actually like look down upon others for using it. You know, these kids are really smart now.”
Even so, Oakwood School is also requiring that more writing is done in-class and it’s evaluating various forms of technology that can be harnessed to limit and prevent AI use, including so-called browser lockdowns on computers. “Kids can still type, but they can’t access other things on their computers,” says Muñoz. She adds that, “we utilize things like Google Docs ‘Version History’ checker to sort of keep an eye that kids are staying within their documents doing the work … because we’re seeing that it’s just so easy to default to using AI.” But Oakwood is not currently using AI checkers to evaluate whether students use the technology in completing assignments. “We actually did a professional development for teachers in a recent meeting where we took an email that I used AI to help me write and we put it through three different AI checkers and they all came up with different responses. One said no AI was used, one said about 50 percent. They all said different things. So we actually use that as an example for our teachers to say, ‘We can’t actually trust AI against AI.’
“And if anything,” she continues, “it’s just gonna build distrust with our students. That’s not a direction we’re choosing to go in because there’s really no proof that those work. And if anything, it’s just gonna build more distrust.”
School leaders and students are also navigating a host of other concerns around AI from thinking about whether AI will increase or diminish career prospects in certain fields (like graphic design or video game design) to looking at the environmental impact of the technology, which is dramatically increasing energy usage across the globe.” At a school like Oakwood, sustainability is really important to our students and our community,” says Muñoz. “They are thinking about the long-term effects of constant AI use on our planet.”
For parents and tech entrepreneurs Adam Ayers and Victoria de la Fuente, who live in Los Angeles and have two young boys, the dawning of generative AI factors into their decision making around schooling. The couple (who founded social platform Doomscrollr) currently send their elder son to a Montessori school “because they’re really focusing on independence, on self-learning and teachable life skills,” says de la Fuente, who is also the founder of Zillion Trillion, an education-focused Substack newsletter and podcast. More than ever, the couple want to avoid schools that put a focus on memorization, while favoring schools that lean into process-based learning. “From what I can see in the schools I’ve been visiting, they’re still very much pushing the memorization route. It just feels like in the age of AI where they can just have vast access to all that information. It feels irrelevant,” says de la Fuente.
That’s something that resonates with Muñoz: “For us, it’s about finding a balance of when can we insert AI appropriately without losing that human piece. It’s really pushed us to work with teachers on how we can reevaluate the way we are teaching and maybe change some of our assessments a little bit to focus on more of the human elements that AI can’t replicate. How do we really incentivize being human? That’s a big part of our conversation.”
