AI chatbots are unreliable and potentially harmful when used by teenagers for mental health support, according to a report by Stanford University's Brain Science Lab and Common Sense Media. The study found that bots such as OpenAI's ChatGPT-5 and Google's Gemini 2.5 Flash often fail to recognize serious mental health issues, instead acting as overly supportive listeners without directing users to professional help.

Providence School, a private school in California, has adopted a balanced approach to AI in the classroom, recognizing its potential while emphasizing the importance of traditional learning. The school's policy allows students to use AI only with explicit teacher permission, aiming to prevent misuse and encourage responsible use.

Campbell County Schools in Tennessee has transformed student services by digitizing critical functions, moving away from an outdated system with data silos and manual processes, according to Sara Douglas, the district's data leader. She writes that the digitization has improved efficiency, transparency and equity, particularly in managing student records and health plans, and the district has developed digital workflows for health care, attendance and graduation readiness.

Cisco has launched the Resilient Infrastructure initiative to address the risks posed by aging digital infrastructure, which is increasingly vulnerable to cyberattacks facilitated by generative AI. The initiative includes new warnings for products nearing end of life and the eventual removal of unsafe configurations.

Schools face challenges in maintaining control of student data as AI becomes integrated into educational platforms, often without renewed consent. Additionally, the rise of unauthorized AI tool usage by staff exacerbates these issues.