OpenAI Introduces MentalHealthBench for AI Mental Health Responses
OpenAI released MentalHealthBench, an expert-informed benchmark designed to evaluate AI responses across realistic mental health conversations. The benchmark assesses both helpfulness and safety of AI replies in these sensitive interactions, providing a standardized way to measure model performance in mental health contexts.
Sources and evidence
Summary last validated Sep 28, 2026
Reader actions
Report an issue
Use this for an incorrect summary, wrong source, duplicate story, or wrong category. Submissions are private and do not change the story automatically.
Related coverage
- industry
Oracle uses ChatGPT and Codex to speed recruiting, engineering, and operations workflows
OpenAI published a customer story describing how Oracle applies ChatGPT Work and Codex across recruiting, engineering, and operations. According to the post, Oracle uses these tools to turn specialist knowledge into fast, repeatable workflows, compressing tasks that previously took days into minutes.
- industry
LegalOn cuts Codex costs 65% with task-matched model routing
LegalOn reduced estimated daily Codex costs by 65% while maintaining development speed, according to OpenAI. The legal tech company matched its Astra, Sol, and Luna models to specific tasks and managed budgets strategically, cutting spend without slowing engineering output.
- industry
Pollo AI uses GPT-5.6, GPT-6 Astra, and GPT-Image-2.5 for ad creation
Pollo AI is using OpenAI's GPT-5.6, GPT-6 Astra, and GPT-Image-2.5 to help creators turn ideas into detailed images and cinematic video ads, according to OpenAI's blog. The post describes how the platform supports campaign creation from concept to finished creative.
- research
NIST Study Maps How Toxic Adulterants in Fentanyl Vary Across US
NIST researchers analyzed fentanyl samples nationwide and found that the toxic substances mixed into the drug vary by region and shift over time. The study aims to help first responders and law enforcement identify dangerous adulterants in local supplies, which the agency says can help save lives.
- research
Apple Introduces Normalizing Trajectory Models for Few-Step Generation
Apple researchers introduced Normalizing Trajectory Models (NTM), which model each reverse diffusion step as an expressive conditional normalizing flow with exact likelihood training. NTM combines shallow invertible blocks per step with a deep parallel architecture, addressing few-step generation without sacrificing the likelihood framework that distillation, consistency training, or adversarial methods discard.