2 min readEnglish
Claude Fable 5.2 arrives with lower prices, plus Anthropic's IPO roadshow
This week brings a major new Claude model release with lower prices, plus notable moves from Anthropic and other AI labs. Here is what matters if you're using AI tools at work.
Claude Platform
Claude Fable 5.2 launches, cheaper and built for complex, long-running tasks
Anthropic released Claude Fable 5.2, a new model designed for demanding reasoning and long, multi-step agentic work, with a 1M token context window. It costs $8 per million input tokens and $40 per million output tokens, down from $10 and $50 for the previous version, so teams get more capability for less money. A restricted version, Claude Mythos 5.2, offers the same power only to organizations that go through Anthropic's verification programs.
Anthropic
Anthropic launches roadshow for its planned stock market debut
Anthropic began the roadshow for its proposed initial public offering of Class A common stock. This is an early step toward the company going public, which can affect its funding, governance, and long-term priorities as a major AI provider.
claude.dev
Anthropic tests safety controls with a 1,000-agent team running Claude Fable 5.2
Anthropic ran an experiment with 1,000 AI agents powered by an early version of Claude Fable 5.2, working together with shared tools and a common database under layered safety controls. The team completed a complex formal proof task, showing how larger organizations might coordinate many agents safely as adoption of multi-agent AI grows.
OpenAI
ChatGPT Enterprise now easier to buy for companies already using AWS
ChatGPT Enterprise is now available through AWS Marketplace, letting organizations purchase and bill it through their existing AWS account instead of setting up a separate contract. This simplifies procurement for companies that already manage other software purchases through AWS.
OpenAI adds business profiles inside ChatGPT
OpenAI introduced a new feature letting companies manage their business information inside ChatGPT, see insights on how their business appears to users, and convert user interest into action. This gives businesses a new channel to manage their presence as more people search and ask questions through ChatGPT.
OpenAI and Upwork research shows AI is reshaping freelance work and budgets
New research from OpenAI and Upwork found shifting demand for freelance skills, new combinations of abilities becoming valuable, and larger budgets and more hours going toward AI-related projects. This matters for managers who hire freelancers or plan project budgets involving AI work.
Google DeepMind
Google DeepMind releases Gemini 4 Carbon
Google DeepMind announced that Gemini 4 Carbon is now available, a new model in its Gemini line. Teams evaluating AI providers alongside Claude or ChatGPT may want to check how it compares for their use cases.
Claude Code
Claude Code update is for developers only
Claude Code, Anthropic's coding tool, released version 2.1.322 adding a faster processing mode on Google's Vertex platform in certain regions. This is a technical change relevant only to software development teams using Claude Code.
Summary prepared by Wisly AI from the official sources linked above. If something affects you, open the source before deciding.