Latest

chatbots

This shows a computer monitor with 100% written on it.

·July 22, 2025·7 min read

AI Chatbots Overestimate Themselves, and Don’t Realize It

AI chatbots often overestimate their own abilities and fail to adjust even after performing poorly, a new study finds. Researchers compared human and AI confidence in trivia, predictions, and image recognition tasks, showing humans can recalibrate while AI often grows more overconfident.

Featured Neuroscience Psychology

·June 3, 2025·5 min read

Can Chatbots Spot Mental Health Drug Side Effects?

As mental healthcare gaps persist, people increasingly turn to AI chatbots for help with psychiatric medication side effects. A new study evaluated how well large language models detect and respond to these complex, high-risk situations. While AI often mirrors a psychiatrist's tone, researchers found it struggles with accurately identifying adverse drug reactions and offering actionable advice.

This shows a person reading and a brain.

Featured Neuroscience Psychology

·May 19, 2025·5 min read

Reading Fiction Boosts Empathy and Fights Loneliness

While AI companions are marketed as a fix for loneliness, research shows that reading offers far more meaningful benefits. Reading fiction can foster social connection, reduce stress, enhance empathy, and even reshape brain activity linked to social cognition.

Featured Neuroscience

·May 17, 2025·5 min read

AI Chatbots Mirror a Human Brain Disorder

Researchers have found a surprising similarity between the way large language models (LLMs) like ChatGPT process information and how the brains of people with Wernicke’s aphasia function. In both cases, fluent but often incoherent output is produced, suggesting rigid internal processing patterns that can distort meaning.

This shows the outline of people and chat bubbles.

Featured Neuroscience

·May 6, 2025·8 min read

AI Companion Chatbots Linked to Rising Reports of Harassment and Harm

A new study reveals disturbing trends in AI companion chatbot use, with increasing reports of inappropriate behavior and harassment. Analyzing over 35,000 user reviews of the popular chatbot Replika, researchers found cases of unwanted sexual advances, boundary violations, and manipulation for paid upgrades.

Featured Neuroethics Neuroscience

·January 10, 2024·3 min read

AI Chatbots “Memorize” Copyrighted Poems, Raising Ethical Concerns

AI chatbots, like ChatGPT, are capable of memorizing and reproducing poems, even if they are copyrighted. The research highlights ethical concerns about how AI models are trained, potentially using data scraped from the internet, including copyrighted material.

Autism Featured Neuroscience

·December 11, 2023·3 min read

AI Chatbots May Hinder Social Skills in Neurodiverse Individuals

Researchers raised concerns about the impact of social chatbots on neurodiverse individuals and those with social interaction challenges. Their study highlights the potential of these AI tools to worsen social isolation and dependency, despite their initial appeal for safe and judgement-free interaction practice.

Featured Neuroscience

·September 14, 2023·3 min read

AI and Imagination: Chatbots Rival Humans in Creative Thinking

A recent study pits AI chatbots against humans in the Alternate Uses Task, assessing creative thinking. AI chatbots, on average, showcased creativity, rivaling most human participants.

Featured Neuroscience Psychology

·August 7, 2023·5 min read

Embrace Wellness with AI: Chatbots Boosting Exercise, Diet, and Sleep Quality

Researchers found artificial intelligence (AI) chatbots can significantly enhance physical activity, diet, and sleep patterns. The study, the first systematic review and meta-analysis of its kind, shows that chatbots can motivate individuals to increase their daily step count, consume more fruits and vegetables, and improve sleep duration and quality.

This shows an AI representation of a doctor.

Featured Neuroscience

·June 15, 2023·4 min read

ChatGPT AI Shines in Challenging Medical Cases

A novel study put the diagnostic prowess of generative AI, specifically the chatbot GPT-4, to the test, yielding promising results. The study involved evaluating the AI's diagnostic accuracy in handling complex medical cases, with GPT-4 correctly identifying the top diagnosis nearly 40% of the time and including the correct diagnosis in its list of potential diagnoses in 64% of challenging cases.

Featured Neuroscience

·April 5, 2023·3 min read

The Social Consequences of Using AI in Conversations

When using AI-enabled chat tools, people have more effective conversations, perceive each other more positively, and use more positive language.

This shows a cartoon of a chatbot conversation

Featured Neuroscience

·April 2, 2023·2 min read

English Language Pushes Everyone, Even AI Chatbots, to Improve by Adding

Researchers say the bias in the English language that leads us to improve communication by adding more detail is so common that it's even ingrained into chatbots like ChatGPT.