ChatGPT Dominates the Generative AI Field, Capturing 60% Market Share


OpenAI CEO Sam Altman proposed a three-stage theory for AI product development: first, chat models (e.g., ChatGPT); second, agent-based systems (e.g., Codex); third, proactive AI that runs continuously and automates tasks. The current agent stage is the largest product category, primarily responding to customer needs, yet users still face limitations.....
xAI released the preview version of Grok Imagine Video 1.5, entering the AI video generation market. The model can convert a single static image into a short video, supporting output at 720p resolution. After users upload an image, they can describe the camera movement, visual rhythm, and atmosphere through text prompts. The model retains the original image details, lighting, and style, generating a natural and smooth dynamic video.
OpenAI has significantly upgraded ChatGPT's memory function with a new system based on Dreaming V3, addressing outdated and inaccurate memory issues while enhancing scalability. It moves beyond strong prompting, enabling intelligent evolution without relying solely on explicit user instructions.....
OpenAI announced a major upgrade to ChatGPT's memory system, addressing two key pain points: 'outdated information' and 'information bias'. The new architecture improves the accuracy of information while reducing computation costs to one-fifth of the original. This feature was first launched in April 2024, and this upgrade represents a core evolution in the underlying technology.
OpenAI recently upgraded ChatGPT's mobile and web versions to enhance conversation naturalness and efficiency. Key updates include a hidden gesture on mobile: long-pressing the send arrow to access an intelligence level selector, allowing users to control model computation and switch response tiers for flexible intelligence adjustment.....