A benchmark called OSWorld-Verified, designed to monitor AI's ability to navigate desktop environments, found that GPT-5.4 scored 75%, up from 47.3% with its GPT 5.2 model. That also beats the average ...

GPT-5.4 is also more reliable, producing 18% fewer errors and 33% fewer false claims than GPT-5.2, according to OpenAI.

Anthropic’s CEO discusses his company’s $10 billion raise, the fields that are most at risk of AI-driven employment disruption and the state of safety in AI development. Photo: Maurizio Martorana for ...

Your AI assistant, now an ad feed. When you purchase through links on our site, we may earn an affiliate commission. Here’s how it works. Yesterday, OpenAI formally announced that it will begin ...

Google’s own ‘code red’ response to ChatGPT has started paying off. Google’s own ‘code red’ response to ChatGPT has started paying off. is a London-based reporter at The Verge covering all things AI ...

Sam Altman issued a "code red" memo directing OpenAI to prioritize ChatGPT quality. The company is delaying advertising initiatives. Google’s Gemini 3 has recently scored higher than ChatGPT on ...

This voice experience is generated by AI. Learn more. This voice experience is generated by AI. Learn more. A few days ago, on November 26th, right before Thanksgiving, OpenAI, the maker of ChatGPT, ...

OpenAI has informed its application programming interface (API) users that some of their data may have been exposed in a breach impacting Mixpanel, a data analytics supplier. In a November 26 blog ...

OpenAI has sent out emails notifying API customers that its chatgpt-4o-latest model will be retired from the developer platform in mid-February 2026,. Access to the model is scheduled to end on ...

Nov 10 (Reuters) - OpenAI is weighing building consumer health products, including a generative AI-powered personal health assistant, as the ChatGPT maker aims to move beyond its core offerings, ...

Threat actors are weaponizing the OpenAI Assistants Application Programming Interface (API) to deploy a backdoor and manage compromised devices remotely. The backdoor was discovered by Microsoft ...

Microsoft uncovers a months-long campaign where threat actors used OpenAI’s legitimate API as a covert command-and-control channel, bypassing traditional detection methods. In a newly uncovered ...