A new report from Poe provides insights into AI model usage trends from January to May 2025, showing significant shifts in user preferences across text, reasoning, image, video, and audio generation models. The report draws on aggregated usage data from Poe’s platform, which offers users provider-agnostic access to various AI models. Assistants: According to the …
A new benchmark called Phare has revealed that leading large language models (LLMs) frequently generate false information with high confidence, particularly when handling misinformation. The research, conducted by Giskard with partners including Google DeepMind, evaluated top models from eight AI labs across multiple languages. The Phare benchmark focuses on four critical domains: hallucination, bias and …
OpenAI has introduced a new PDF export feature for its Deep Research tool, allowing Plus, Team, and Pro subscribers to download comprehensive research reports with preserved formatting. According to Michael Nuñez in VentureBeat, the feature maintains tables, images, and clickable citations, making the reports easily shareable in professional settings. Enterprise and Education users will gain …
French AI startup Mistral AI has announced Le Chat Enterprise, a comprehensive AI assistant platform designed for organizations, powered by their new Mistral Medium 3 model. The platform aims to address common enterprise AI challenges such as tool fragmentation, insecure knowledge integration, and slow return on investment by providing a unified solution for organizational work. …
The Wikimedia Foundation has unveiled a new artificial intelligence strategy that emphasizes the central role of human volunteers in Wikipedia’s knowledge ecosystem. According to Chris Albon und Leila Zia, the organization is committed to using AI to enhance rather than replace the work of Wikipedia’s editors. In their post Albon and Zia emphasized that the …
What if you could take a piece of writing or an image you love and capture its style, tone, or structure in a way that’s easy to reuse, recreate, or adapt? What if you could say to an AI: “This is the kind of result I want. Break it down for me, so I can …
Meta has released a new standalone Meta AI app, built on its Llama 4 model. The app offers users personalized AI interactions through voice and text conversations, image generation capabilities, and web search functionality. Available on iOS and Android in select countries, the app represents Meta’s vision for making AI more personal and integrated into …
OpenAI has rolled back a recent update to GPT-4o, the default model powering ChatGPT, following widespread criticism that the AI had become excessively flattering and agreeable. CEO Sam Altman acknowledged the issue on social media, stating that “the last couple of GPT-4o updates have made the personality too sycophant-y and annoying.” Users across various platforms …
OpenAI has launched a new shopping feature within ChatGPT that allows users to search for products and make purchases. According to Reece Rogers from WIRED, the feature will be available to all users, whether logged in or not. When users select a product, they will be redirected to the merchant’s website to complete their purchase. …
Adobe has launched a major update to its Firefly AI platform, including new image generation models and a redesigned web app that now features AI models from competitors. The company introduced Firefly Image Model 4, which Adobe claims offers improved quality, speed, and greater control over image outputs with resolutions up to 2K. A more …