Following a teenager’s suicide found to have been provoked by an AI chatbot, people have raised concern about the role of artificial intelligence (AI) in teen mental health.

Stories about mental health issues with AI tend to happen to people who are already struggling emotionally, and people affected may lack clear judgment. According to Adam Tachner, an expert on the corporate side of AI, AI was created as an augmentation tool to work alongside clear-thinking humans to make their lives easier, not as a replacement for thought. Tachner previously served as head of legal, finance and corporate development for Groq AI.

“It’s this notion of augmentation,” Tachner said. “Would I trust the tool to help me without sacrificing my own independent intelligence and my ability to discern good advice from bad advice? I’m happy to have another tool in my toolbox. I’m not interested in giving over control of my life.”

Other people may think of AI more as a search engine than a more diverse tool.

“[AI] is basically the same as looking something up on Google,” freshman Theo Marks said. “It’s just on a different website.”

Current AI software has many capabilities beyond just answering questions, and it works differently from just searching on Google. According to Tachner, large language models take input from a user and run statistical analysis to predict a likely next word, hence the name generative AI. Many experts warn this algorithmic nature of AI could make it dangerous.

“AI is so new, and we’re still learning about the benefits and challenges of AI,” life skills and social studies teacher Blaine Dzwonczyk said. “I think that given that there’s no human being on the other side, there’s a possibility for things to go really wrong.”

The situation surrounding AI has changed significantly in recent years. AI chatbots have supposedly encouraged teen suicides, and according to a survey from the Center for Democracy and Technology (CDT), one-fifth of the 1,000 high school students surveyed reported having had or knowing someone who has had a romantic relationship with AI.

“I think it would raise a lot of questions about humanity if AI were to replace social interactions,” Dzwonczyk said.

For many, the question becomes where to draw the line. Some fields could require too much confidence in AI. There may be a grey zone. For example, teens may lack access to a human therapist and so turn to AI instead.

“On one hand, of course, if someone is struggling [with mental health] I want them to get whatever support they can access,” Dzwonczyk said. “On the other hand, I’m concerned about the possible danger of AI.”

Using AI with the concept of augmentation in mind doesn’t make it completely safe either.

“[Reliance is] one form of danger that comes from the integration of AI into our lives,” Tachner said. “[However,] I don’t think complete reliance is necessary for there to be problems with AI.”

According to Dzwoncyzk there is no obvious answer to the question of how to use AI safely individually without guideline from companies.

“I am concerned,” Dzwonczyk said. “I wonder if the companies that create AI are really well-suited to be regulating their own product.”

With approximately 20% of teens struggling with mental health, there is a lack of therapeutic services. For many teens AI is a new solution to this problem. President Trump’s current political administration leaves it up to parent companies to regulate their own products. Tachner noticed that when companies prioritize maximizing profits,  AI can be dangerous.

“[AI companies] can do a tremendous amount of damage to their users and societally,” Tachner said. “So I would hope that they would do more than the minimum necessary [to protect users], but I don’t trust them to do more.”

This article was updated on Nov. 7  for clarity and factual accuracy.