
OpenAI announces the acquisition of Promptfoo, a cybersecurity startup specializing in testing and validating AI systems. The technology will be incorporated into the Frontier platform for AI agents, while Promptfoo's open-source project will continue to be maintained. The purchase reflects OpenAI's strategy to consolidate infrastructure for the development and operation of autonomous agents, with a focus on security. Developers should monitor the future of the open-source project.

Cursor Composer 2's performance against a more expensive model like Opus 4.6 points to a significant shift: the AI race for development is no longer just about maximum capability, but about cost per useful result.

Embeddings are the technical foundation for making RAG (AI with your data) work, transforming text into numbers that represent meaning. Without them, your AI is dumb or too expensive. This practical guide with Python and sentence-transformers shows how to implement semantic search, solve synonym problems, and understand why they're essential for building intelligent and efficient chatbots.
Anthropic has released Sonnet 4.6, the new daily "workhorse." This model offers cutting-edge performance in coding and agentic workflows, nearly matching Opus 4.6. Highlights include 1 million context tokens (beta), new "Effort and Thought" controls for granular reasoning, and GA features like Code Execution and Web Search. It's the ideal choice for serious engineering, combining intelligence and cost-effectiveness.

OpenAI launched two new models today — and they are not for you to use directly in chat. They are for delegation. The era of agents has a new cost logic, and it changes how you will build with AI.