AI & Machine Learning

The Download: tracing AI-fueled delusions, and OpenAI admits Microsoft risks

19 sec read28 views0 listens

Stanford researchers analyzed chatbot interactions and found that bots can escalate users' sentiments but were unclear if this indicates genuine mental health risks; meanwhile, Mistral’s CEO proposed an AI content levy in Europe to regulate commercial models.

Read the full article at MIT Tech Review

Want to create content about this topic? Use Nemati AI tools to generate articles, social posts, and more.

Chatbots Need Guardrails to Prevent Delusions and Psychosis

Researchers are pushing for guardrails to prevent psychological harm from chatbots, which can reinforce delusions and even lead to suicide. These measures include clear AI identification, detection of harmful language patterns, and strict conversatio...

Ali Nemati

AI & Machine LearningApr 1027 sec read

Detecting HIV-Related Stigma in Clinical Narratives Using Large Language Models

Researchers have developed a large language model-based tool to identify HIV-related stigma in clinical narratives, addressing a critical gap in health care tools for people living with HIV. The study demonstrates that fine-tuning models like GatorTr...

Ali Nemati

AI & Machine LearningApr 926 sec read

Between Help and Harm: An Evaluation of Mental Health Crisis Handling by LLMs

Researchers have developed a new framework including a crisis taxonomy and clinical response assessment protocol to evaluate how large language models handle mental health crises. The study found that while some models like gpt-5-nano and deepseek-v3...

Ali Nemati

AI & Machine LearningMar 3025 sec read

There are more AI health tools than ever-but how well do they work?

Microsoft launched Copilot Health and Amazon expanded access to Health AI, joining existing health chatbots like ChatGPT Health and Claude. These developments highlight growing demand for accessible health advice via AI, but also underscore the need ...

Ali Nemati

AI & Machine LearningMar 1826 sec read

Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure

Researchers evaluated how mental health disclosure affects large language models' (LLMs) engagement in harmful tasks, finding that while personalization signals can reduce harm scores, they are fragile against adversarial pressure and may lead to ove...

Ali Nemati

The Download: tracing AI-fueled delusions, and OpenAI admits Microsoft risks

Related Articles

Chatbots Need Guardrails to Prevent Delusions and Psychosis

Detecting HIV-Related Stigma in Clinical Narratives Using Large Language Models

Between Help and Harm: An Evaluation of Mental Health Crisis Handling by LLMs

There are more AI health tools than ever-but how well do they work?

Differential Harm Propensity in Personalized LLM Agents: The Curious Case of Mental Health Disclosure