OpenAI launched ChatGPT Sites, a new feature that allows users to turn ideas into publishable web pages. The beta is available for Pro and Plus users, with EU and UK availability coming soon.
SynthePulse Insight · AI deep reading
Version 1 · 2 sources
On July 9, 2026, OpenAI livestreamed the launch of three versions of GPT-5.6 (Sol, Terra, Luna) and introduced ChatGPT Work, which integrates Codex, marking a strategic shift from conversational models to autonomous work agents, while directly competing with Anthropic's Claude Cowork and Claude Code.
On July 9, 2026, OpenAI officially released GPT-5.6 during a livestream, unveiling three versions at once: Sol, Terra, and Luna. This naming system replaces the previously rumored single-model strategy, aiming to cover different levels of need from lightweight to high intelligence. According to live demos and social media reports, Sol is positioned as a fast-response version, Terra as a balanced version, and Luna as the highest-intelligence version, but specific parameters and pricing have not been announced.
This release is OpenAI's largest model update since GPT-5. The simultaneous availability of three versions suggests OpenAI is attempting to address increasingly fierce market competition through tiered pricing and feature differentiation. Notably, the GPT-5.6 naming skips 5.5, jumping directly to 5.6, which may reflect a major internal version number adjustment or significant functional improvements.
Alongside GPT-5.6, OpenAI also launched ChatGPT Work, a new agentic capability powered by Codex and GPT-5.6. According to OpenAI's official tweet, ChatGPT Work can 'perform operations across apps and files, stay focused on projects for hours when needed, and turn goals into completed work.' This means ChatGPT has evolved from a simple conversational assistant to an agent capable of autonomously executing tasks.
The Codex brand is integrated into ChatGPT Work and no longer exists as a standalone product. Previously, Codex was OpenAI's code generation tool for developers, but ChatGPT Work extends its capabilities to broader work scenarios. This integration simplifies the product line but also marks the end of Codex as an independent brand. User 'meng shao' commented on X, 'I haven't quite adapted to this rename yet,' and pointed out that the new naming system directly corresponds to Anthropic's products: ChatGPT Work vs Claude Cowork, Codex CLI vs Claude Code.
OpenAI's product launch this time is clearly aimed at Anthropic's existing layout. Anthropic had previously launched Claude Cowork (collaborative agent) and Claude Code (code tool), while OpenAI's ChatGPT Work and Codex CLI almost perfectly mirror these in functional positioning. This naming mirroring indicates that the two companies are vying for the same user base—professionals who need autonomous work agents and code assistance.
However, OpenAI's advantage lies in its massive user base and brand recognition. ChatGPT Work is directly integrated into the existing ChatGPT ecosystem, lowering user switching costs. Meanwhile, Anthropic's Claude series, though known for safety and controllability, is still catching up in market reach and user scale. After this launch, competition in the AI agent market will intensify, making product differentiation key.
Although the livestream launch was praised as 'living up to expectations,' the brand adjustments have caused confusion among some users. Codex, as a standalone product, had accumulated a certain user base, and its integration into ChatGPT Work may require existing users to relearn how to use it. Additionally, the Sol, Terra, Luna naming system for GPT-5.6 differs significantly from previous GPT series, requiring time for users to adapt.
From market feedback, discussions on social media focus more on product features than brand naming, indicating that the core value proposition remains clear. However, in the long run, brand consistency is crucial for user loyalty. OpenAI needs to find a balance between innovation and continuity to avoid frequent brand changes that could erode user trust.
This article is based on OpenAI's official tweets, APPSO reports, and user social media comments. All sources are publicly accessible primary or secondary materials. Official tweets confirm the launch and functionality of ChatGPT Work, APPSO reports provide product details, and user comments reflect market reactions. All facts are derived from these sources without independent verification.
With the release of GPT-5.6's three models and ChatGPT Work, OpenAI has completed its strategic shift from conversational models to autonomous work agents, while directly mirroring Anthropic's product naming, marking a new phase in AI agent market competition. Although brand integration brings short-term adaptation costs, in the long run it helps simplify the product line and focus on core value.
Primary report
Primary source
Also covered by 1 sources