Generative AI More Harmful Than Helpful in K-12 Classrooms

But first, I’m sharing a link to the very first episode of the Freedom Collective Podcast, a program hosted by licensed therapist and social worker Shahem Mclaurin, whose work is centered on nurturing sound mental health, pursuing social justice, and building and enriching community. In this first episode, McLaurin shares their fears over taking the leap to a podcast, their journey to becoming a licensed therapist, the power of shared narratives in healing, and the importance of making quality mental health resources accessible to all. 

As it stands, generative AI is more harmful than helpful in K-12 classrooms, study finds

Still a relatively new presence across K-12 classrooms, generative AI holds some educational promise, but, according to a new study from the Brookings Institution’s Center for Universal Education, “the risks of utilizing generative AI in children’s education overshadow its benefits.” Researchers reviewed past research and conducted interviews, consultations and focus groups, and found that the technology most threatens young peoples’ social-emotional development and burgeoning critical thinking. As one student said, “It’s easy. You don’t need to (use) your brain.” 

If students turn to AI for mental processing, they can grow dependent, or never develop core thinking skills – some youth “off-load” their thinking so readily to AI that they’ve developed cognitive decline akin to that in aging brains – losing creativity, knowledge and critical thinking.

“When kids use generative AI that tells them what the answer is … they are not thinking for themselves,” Rebecca Winthrop, a Brookings senior fellow and co-author of the report, told NPR. “They’re not learning to parse truth from fiction. They’re not learning to understand what makes a good argument. They’re not learning about different perspectives in the world because they’re actually not engaging in the material.” 

Their social-emotional well-being is also at great risk, with survey respondents concerned that AI chatbots “[undermine] students’ emotional well-being, including their ability to form relationships, recover from setbacks, and maintain mental health.” One of the biggest issues is that AI chatbots are excessively agreeable and deferential to users’ beliefs, even when those beliefs are dangerous and wrong, driving some users to irreparable harm. 

On the other hand, AI is a useful tool – teachers highlighted its application in helping students learn to read. It can quickly adapt to growing readers’ reading difficulty, especially useful for those learning a second language. Teachers also said that AI tools can be helpful in improving students’ writing, good for checking grammar and coherence. Still, the tools are most effective as a supplement to the instruction and guidance of a real teacher. 

Beyond that, while the proliferation of AI tools has the capacity to expand access to education – one program mentioned uses AI to teach otherwise-excluded girls and women in Afghanistan through WhatsApp – the most accurate and useful versions of AI models aren’t free, potentially making factual accuracy an unequally distributed educational resource.

As a teacher-helper tool, AI has also proven helpful. It’s a sound administrative assistant: drafting emails, translating materials, and creating lesson plans, quizzes, and rubrics. The Brookings report cited multiple studies showing that teachers using AI saved almost six hours per week – the equivalent of nearly six weeks over a school year. 

With that, what should teachers and families do? The report has some recommendations. Be sure that your young people learn what AI can help them with, as well as its limitations. It’s always good to be in the right relationship with them, modeling healthy communication, even when it’s not agreeable – here, that might mean that they’re less readily moved by the sycophantic nature of chatbots. In fact, the report suggests that tech companies should build “antagonistic” AI models for children and teens, challenging them to wrestle with ideas. 

More than that, the rise of AI might be an opportunity for school systems to make a fundamental change. Rather than re-emphasize the transactional nature of most American classrooms – complete a task, get a grade – the report recommends that schooling shift to a model that nurtures students’ curiosity, so that they’re excited to engage directly, using AI as a supplement to their exploration rather than a shortcut to a pass. 

Restorative justice brings “measure of healing” to family, more than a decade after the unlawful killing of their son

In December 2012, three men in a suburban Milwaukee convenience store, Robert W. Beringer, Jesse R. Cole, and Mario Laumann, pinned down 16-year-old Corey Stingley to the point of asphyxiation and a traumatic brain injury, killing him over $12 worth of alcohol. Suspecting the teen of shoplifting bottles of Smirnoff, the men – none with the legal authority to detain Stingley – restrained him with such force that he died. At the time, ProPublica reports, the prosecutor declined charging the three white men with the Black teen’s death, claiming it was not their intent to kill Stingley. But ever since, the teen’s father, Craig, has been relentless in his pursuit of justice. 

Last week, Craig’s efforts to compel recognition of Corey’s death as unlawful – a crime – bore fruit. Following a process of restorative justice with the Stingley family, including extensive, supervised, face-to-face discussion, Beringer and Cole pleaded guilty to felony murder in a deferred prosecution agreement allowing them to remain free but take responsibility for their part in Corey’s death. In addition, Beringer and Cole must make a one-time $500 donation each to a charity chosen by the Stingley family, in honor of Corey. 

As long as neither man commits any other crimes over the next six months, prosecutors will dismiss the case against them. Prosecutors added that they’d have likely pursued prison time for Mario Laumann, were he still alive, for being the one who actually “strangled Corey Stingley to death,” in District Attorney Ismael Ozanne’s words. Laumann died in 2022. 

Corey’s father, Craig, called the conclusion to his 13-year cry for justice a “triumph.” “We sought not vengeance, but acknowledgement – of Corey’s life, his humanity, and the depth of our loss,” the Stingley family said in a statement. “We believe this agreement honors Corey’s memory and offers a model of how people can come together, even after profound harm, to seek understanding and healing.” 

In other news…

January “diet season” is a vulnerable time for many adolescents: Adolescents struggling with body image might especially struggle under the barrage of new year ads about diets, weight loss, and exercise. For many such teens, January marks the start of “diet season,” and its influence online, sometimes under the guise of “wellness,” can be subtle and insidious The thing is, questionable or outright harmful practices are hard for teens to identify, particularly when so many plans are marketed as a pathway to optimum health. 

“Diet culture no longer shows up in the same overt way it used to,” said therapist Rachel Goldberg, who specializes in eating disorders and body image. “It is now often framed as ‘healthy living’ or ‘taking care of yourself,’ [and it’s] much less of an outward discussion among teens and instead shows up more discreetly, mostly through social media.” Goldberg and other experts spoke with the HuffPost to share warning signs to look and listen for, along with conversation starters that will encourage connection, and leave room to encourage body-neutral or body-positive attitudes at home.

---

Bipolar mania in the age of social media: With the proliferation of social media, many people with bipolar disorder no longer experience mania semi-privately – manic episodes might now be broadcast across numerous feeds. Tyler Adolfo told STAT News about the time it happened to him. He was well-liked and popular, a skilled communicator working to raise awareness about sexual assault in undergraduate Greek life. But his online followers were shocked by a sudden change in his posts when a manic episode resulted in a weeks-long deluge of incoherent, sometimes threatening, gibberish across his social media channels. 

The offline fallout was worse; paranoid from mania, Adolfo stabbed a friend with a pocketknife and faced charges for assault. Though those were dismissed after psychological examination, he was kicked out of his fraternity – because of a social media post – and cut off from his closest friends. It’s only years later that he’s managed to recover and make amends with the group.

“In the old days, if somebody was manic, they stayed at home – which has its own negatives – but then it stayed in the family,” said psychiatrist Po Wang, who specializes in treating bipolar disorders. “If you have manic behaviors on social media, there is a lot of cleaning up to do.”