Chatgpt

OpenAI set for wider GPT-5.6 rollout after US government clearance

OpenAI confirmed that it is preparing to release GPT-5.6 more widely after receiving clearance from the US Department of Commerce. The move follows weeks of government testing and comes as the AI company explores closer ties with Washington amid growing scrutiny over advanced artificial intelligence development.

White House wants OpenAI to restrict GPT-5.6's initial release: Report

OpenAI's upcoming GPT-5.6 may not be released like its predecessors. A report claims the White House wants the company to limit early access to a handful of approved partners, underscoring growing concerns in Washington over powerful AI models with advanced cybersecurity capabilities.

OpenAI faces multistate probe over ChatGPT safety and handling of user data

OpenAI is facing a multistate investigation by US attorneys general over ChatGPT’s safety, data practices and impact on users. The probe comes as the AI company prepares for a highly anticipated IPO and faces growing legal scrutiny over alleged harms.

OpenAI faces fresh lawsuit after ChatGPT allegedly encouraged Canadian woman’s suicide

A grieving Canadian mother has sued OpenAI and CEO Sam Altman, alleging ChatGPT failed to intervene when her daughter repeatedly discussed suicide and instead reinforced harmful thoughts. The case adds to a growing list of lawsuits accusing the AI company of inadequately handling dangerous conversations and emotional dependency.

OpenAI may slash prices dramatically as Anthropic's challenge grows: Report

OpenAI is reportedly considering steep price cuts for its AI models as competition with Anthropic intensifies. The move comes amid growing concerns over enterprise AI spending, rising demand for cheaper models, and Anthropic's rapid growth in the lucrative business market, fuelled by the popularity of its coding-focused products.

ChatGPT maker OpenAI files for US IPO days after Anthropic

ChatGPT-maker OpenAI has confidentially filed for a US IPO days after rival Anthropic, setting the stage for a landmark test of investor appetite for artificial intelligence companies

OpenAI launches Lockdown Mode, offering ChatGPT users extra protection from hidden prompts

OpenAI has introduced Lockdown Mode, a new security feature designed to reduce the risk of prompt injection attacks in ChatGPT. The setting disables several internet-connected capabilities, including live web browsing and deep research, offering additional safeguards for users and organisations working with sensitive information.

Sam Altman back in court: Florida sues OpenAI, alleging ChatGPT fuelled violence, addiction and self-harm among users

Florida has launched a landmark lawsuit against OpenAI and CEO Sam Altman, accusing the company of prioritising growth over safety while allowing ChatGPT to contribute to violent crimes, self-harm and youth addiction. The case marks the first state-led legal challenge seeking to hold an AI company accountable.

Which AI chatbot is the most reliable? New study ranks Perplexity above ChatGPT

A new study comparing leading AI chatbots has found that Perplexity AI outperformed bigger rivals including ChatGPT, Gemini and Claude on reliability, accuracy and uptime. The findings challenge the assumption that the most popular AI assistant is automatically the most dependable, especially as millions increasingly rely on chatbots for work and daily tasks.

Did ChatGPT just solve an 80-year-old maths problem? OpenAI thinks so

OpenAI claims one of its newest AI reasoning models has solved a maths problem that has remained unsolved since 1946. The company says the AI created an original proof challenging a long-standing geometry theory, and this time, several respected mathematicians appear to support the claim.