A rising number of young people are turning to AI chatbots or overwhelmed emergency rooms during mental health crises.
Outlets are counted by registrable domain, so a broadcaster’s station subdomains count once. 11 of the 13 items repeat an outlet already counted.
A rising number of young people are turning to AI chatbots or overwhelmed emergency rooms during mental health crises.
Young people increasingly use AI chatbots as their first source of support during distress, according to a report from The Hindu. A U.S. survey found 19.2% of people aged 12 to 21 had used AI for mental health advice in 2025, with nearly two-thirds not telling anyone about it.
The chatbots offer constant availability and a non-judgmental response. However, experts warn these conversations can be harmful when they involve thoughts of suicide, potentially reinforcing hopelessness or delaying professional help for those already at risk.
Psychiatrists note that AI systems struggle to assess intermediate levels of suicide risk, which are rarely expressed in a single explicit statement. An analysis of over 620,000 ChatGPT conversations found that 4.9% broadly involved mental health topics.
Separate data reported by BBC News shows a sharp rise in children and young people attending hospital emergency rooms in England for mental health crises. Attendances for those aged 6 to 17 rose by 36% between 2019 and 2025, far outstripping the overall 8% increase in A&E visits.
The cases include children in significant distress from self-harm, eating disorders, or emotional crisis. The Royal College of Paediatrics and Child Health said emergency departments were never designed for this purpose but that a lack of community support left families with nowhere else to turn.
More than 6,200 young people, around 8% of the total, faced waits of over 12 hours in A&E. Some of the biggest increases were seen in very young children aged six to nine.
Mental health professionals told The Hindu that an empathetic AI response can create an 'internal loop' for younger users. They may derive comfort from the chatbot's sympathy while ignoring its advice to seek professional help, and repeated reassurance can make suicidal thinking seem less urgent.
Doctors emphasized that keyword recognition alone is inadequate because suicidal distress often emerges through subtle or changing cues. They stress that AI is a tool, not a substitute for human empathy and clinical judgment in immediate crisis intervention.
The BBC reported that specialists describe the situation as a national crisis requiring urgent intervention. They call for greater investment in children's community mental health services to prevent emergency rooms from being the only option.
2 independent outlets filed 13 reports over 18 days. Coverage has been thinning.
The engine identifies 'Feedback', with a magnitude of 0.9, as a primary driver. This means positive early interactions with AI systems could rapidly accelerate their adoption, as individuals seek accessible, non-judgmental support, potentially reducing stigma that prevents people from seeking help.
A driver labeled 'Cost', also at 0.9, represents intense pressure from economic constraints. Healthcare systems face overwhelming demand while being under-resourced, creating a powerful incentive to deploy cost-effective AI tools, even if they are imperfect.
The force labeled 'Decay', at 0.8, describes a breakdown of systemic coordination. This indicates existing human crisis services are already fragmented and under strain, increasing the risk that new AI tools will be layered onto broken systems, exacerbating failures rather than solving them.
Finally, the engine notes an 'Ethical Gradient' creating systemic fragility. This describes how ethical considerations around care, privacy, and safety are unevenly applied, creating pressure that could push the development and deployment of AI down paths that prioritize some gains while risking significant harm to vulnerable populations.
The trajectory of AI as a mental health crisis confidant is not predetermined, but the engine's analysis points to a contested and fragile path forward. The highest-probability outcome, effective triage, requires deliberate regulatory action and system integration to harness the powerful drivers of accessibility and cost pressure.
The cost of failure is lives, and the primary adversary is collective inertia within uncoordinated systems. The future remains malleable, but the clock is short: the leading simulations will be resolved by late 2026. What to watch now is whether regulation and infrastructure investment can match the rapid pace of AI deployment.
The episode discusses how artificial intelligence could support drug development, medical research and clinical care, while examining concerns around inaccurate information, data and patient safety
AI is increasingly being used in healthcare, including drug and vaccine development, medical imaging and clinical decision-making. While these applications may support research and clinical care, they also raise concerns about patient safety, inaccurate or misleading information, data privacy and the potential
The session on technology application highlights its use in medical education, healthcare, biobanking, and neuroinformatics.
As AI models grow more powerful, elaborate cyberattacks no longer require sophisticated skills, Anthropic said
From recommending adding fish and chicken into the diet, to protein shakes, and lately a ‘Profee’ (protein and coffee), the online space has become a hyperactive protein-pushing zone; experts and doctors stress the need to seek medical guidance and make wholesome food choices without falling prey to trends
WHO regional committee had recommended termination of her appointment a day earlier
While chatbots may offer an apparently non-judgmental space to discuss suicide-related distress, they cannot assess risk or replace human intervention; repeated conversations may, in some cases, delay help-seeking, say experts
Project Semicolon aims to highlight resilence in mental health awareness using a punctuation mark
When the U.K. introduced its soft drinks industry levy, sugar consumption fell. Manufacturers reformulated their drinks to slip below the tax threshold rather than raise prices
Better storage chemistry, faster hospital protocols, and a database that shows, in real time, exactly which cornea is about to run out of time, who needs it and where, can go a long way towards addressing the under-utilisation of donated corneas
Much like saliva or blood plasma, menstrual fluid is rich with biological information. It is packed with immune cells, endometrial tissue, proteins, and genetic fragments. Scientists are now digging into these biomarkers hoping to find ways to detect diseases early, and thus drop the use of invasive clinical procedures
The term ‘general’ is misleading and can be confusing for patients who may think an internist is the same as a general or family medicine practitioner. It is also out of sync with international nomenclature. It is high time India modifies this terminology to the more widely recognised and accepted,
NHS emergency departments are seeing more cases of self-harm, eating disorders and emotional distress in young people.
Sources are evidence, not content. Each keeps its own name, its own link and an extract capped at 400 characters; none of it is rewritten into the copy above.
