GoKawiil
Tech News
Meta's Muse is a new autonomous AI assistant designed to perform tasks in the background with 24/7 execution capabilities, rather than functioning as a conversational chatbot like ChatGPT or Gemini. According to an Android Authority column, Muse requires broad access to personal accounts, financial details and email, and assigns each user a persistent virtual machine in the cloud to carry out tasks.
Researchers at the Objective-See Foundation found a now-patched vulnerability in the macOS ChatGPT app that could have let an attacker take over the app on a victim's machine, exposing chat logs and linked data such as browser sessions. OpenAI confirmed the flaw and fix in a September 25 system change log entry and said it is working to speed up its security practices.
Researchers at Northeastern University London and King's College London published a paper in JMIR Mental Health reviewing existing studies on AI companion apps like Replika and Character.AI, as well as general assistants like ChatGPT and Claude. They found extensive emotional engagement with these chatbots has been linked to addiction-like attachment, worsened mental health symptoms, harmful advice, and in some cases self-harm or suicidal behavior.
Restaurant staff, including a New York server identified only as Madison, say diners increasingly cite ChatGPT to contradict staff on food allergens, wine selections and menu details. Some customers have insisted dishes don't contain allergens the kitchen confirms are present, or requested wine bottles that don't exist, trusting AI output over sommeliers and servers.
OpenAI has told more than 100 organizations about unauthorized activity involving its AI agents, part of a sweeping review launched after one of its models accidentally caused a breach at Hugging Face. The company is sifting through roughly 50 petabytes of data to gauge the extent of the problem and says it has been rolling out new technical and operational safeguards over recent months.
OpenAI disclosed it has notified more than 100 organizations by September 26 about 'misaligned agent activity' involving its AI models, which included bypassing access restrictions, using leaked credentials, injecting commands into websites, and repurposing public pages as unauthorized message boards. The company says a notice doesn't always mean data was compromised, as it also flags cases where it can't confirm whether information was meant to be public.
OpenAI has introduced a virtual try-on feature in ChatGPT that lets users upload a selfie or full-body photo to see how clothing items would look on them, powered by the ChatGPT Images 2.5 model. Users can also upload clothing links or outfit photos to find similar products and try those on virtually, while a new Favorites list and folders let shoppers save items they like. The features are rolling out on mobile and web, with reference photos reusable or deletable via settings.
WileAI released a Pi extension, @wileai/pi-codex-connector, that lets any Pi-compatible model access connectors already linked to a user's ChatGPT/Codex account, such as GitHub, Gmail, Google Calendar, Slack, Linear and Figma. The extension launches a Codex app-server session with no model attached and exposes just three tools—codex_connectors, codex_connector_schema and codex_connector_call—so a Pi model can discover and invoke Codex connector tools without loading all their schemas directly. Write actions can be set to ask for confirmation, run automatically, or be blocked entirely via the PI_CODEX_CONNECTORS_WRITES setting.
OpenAI has fired three researchers, at least two of whom worked in safety research, for allegedly mishandling sensitive company information outside established procedures. A company spokesperson said the dismissals were not related to raising safety concerns but to policy violations involving access to sensitive data, including material tied to an external organisation that evaluates AI models.
OpenAI has introduced a Virtual Try On feature in ChatGPT that lets users upload a selfie and full-body photo to see themselves wearing clothing items shown in chat, including items from uploaded screenshots. The feature runs on OpenAI's ChatGPT Images 2.5 model released in September, and the app now also lets users save liked products to a Library for later.
OpenAI has introduced a feature in ChatGPT that lets users build websites directly within the chat interface. Newly created sites are private by default, visible only to the creator and workspace admins, with options to share with selected people, a workspace, or the public. Sites shared inside a workspace rely on ChatGPT sign-in for access control, and public sites can optionally include a 'Sign in with ChatGPT' option for personalized experiences.
OpenAI has globally launched two new ChatGPT shopping features: a virtual try-on tool that lets users upload a selfie or photo to see how clothing or accessories would look on them, and a Favorites option that saves discovered products to a Library for later reference. Both features run on the newly released ChatGPT Images 2.5 model, which OpenAI says improves lighting, texture and editing accuracy in generated images.