Key Takeaways ChatGPT now automatically manages saved memories, eliminating "memory full" errors for Plus and Pro users. All Sora users can now generate 15-second videos on both app and web platforms....
OpenAI has introduced automatic memory management for ChatGPT, addressing a persistent complaint among power users of its AI chatbot.
The feature, which began rolling out to ChatGPT Plus and Pro subscribers on October 15, 2025, automatically prioritizes relevant memories while moving less important information to the background.
The system determines which memories to keep prominent by analyzing factors such as how recently a detail was mentioned and how frequently users discuss particular topics.
Users previously had to manually delete saved information when their memory storage reached capacity, often triggering frustrating "memory full" alerts that disrupted their workflow.
According to OpenAI's official announcement, users retain full control over the feature.
They can turn off automatic memory management, search and sort memories by date, and manually prioritize or deprioritize specific stored information.
The company has also added the ability to view and restore previous versions of saved memories.
The memory management feature works alongside ChatGPT's existing dual-memory system, which includes both "saved memories" that users explicitly request the AI to remember and "chat history" insights that ChatGPT gathers automatically from past conversations.
On October 16, 2025, OpenAI announced significant upgrades to Sora, its AI-powered video generation platform. The updates extend video duration limits and introduce advanced creative tools for professional users.
All Sora users can now create videos up to 15 seconds in length on both the mobile app and web platform, an increase from the previous 10-second default.
For Pro subscribers using the web version, the maximum video length extends to 25 seconds. According to OpenAI's release notes, 15-second videos count as two videos toward daily usage limits, while 25-second videos count as four.
The longer duration allows creators to develop more complete narratives within a single generation, providing enough time to establish tone, build a brief plot, and deliver a conclusion without requiring multiple clips.
Pro users also gained access to Sora's storyboard feature on the web platform.
The tool enables users to plan videos frame by frame, either building scenes from scratch or describing a concept and allowing Sora to generate a detailed storyboard that can be edited.
This feature is designed to give creators more precise control over pacing and narrative structure while reducing the need for multiple generation attempts.
These updates arrive as competition intensifies in the AI video generation space.
Google's Veo platform and other competitors have been advancing their control features, pushing OpenAI to enhance Sora's capabilities beyond simple duration increases.
The storyboard tool represents OpenAI's effort to position Sora as a professional-grade platform suitable for content creators, marketers, and small production teams who need more structured video creation workflows.
Both features are currently available only in the United States and Canada. OpenAI has not announced when these capabilities will expand to other regions, including Europe, where regulatory requirements may delay rollout.
Key takeaways OpenAI has signed a multi-year, $10 billion agreement with AI chipmaker Cerebras Systems to secure computing infrastructure. The deal will deliver 750 megawatts of computing power throug...
KEY TAKEAWAYS: xAI implemented restrictions preventing Grok from editing images of real people in revealing clothing after global backlash California Attorney General Rob Bonta launched investigation...
Key takeaways TSMC reported fourth-quarter net income of NT$505.74 billion, beating analyst expectations and setting a new record. Fourth-quarter revenue reached NT$1.046 trillion, up 20.5% year over...
