GPT-5.5 is reported to hallucinate three times more than the MIT-licensed GLM-5.2 in AI model evaluations.

GPT-5. 5, with an estimated 1-2 trillion parameters, scored an 86% hallucination rate on the AA-Omniscience benchmark.

GitHub developed Qubot, an internal analytics agent powered by Copilot, to enable employees to query data using plain language.

Qubot allows GitHub staff to interact with data without requiring specialized query languages or tools. The project was led by Matteo Vasirani, who manages product analytics and data science at GitHub. This initiative highlights the integration of AI to streamline internal data processes.

The Enterprise-Managed Authorization extension for the Model Context Protocol has reached stable status, allowing centralized MCP server access through an organization's identity provider.

With this extension, users can access all connected MCP servers on their first login without needing per-app OAuth configurations. Organizations like Anthropic, Microsoft, and Okta are adopting this approach. Administrators set access policies once, and the identity provider manages access decisions based on user roles and group memberships, streamlining the process significantly.

GitHub has implemented pull request limits to help manage the volume of contributions in repositories.

This feature aims to reduce noise by controlling the number of pull requests that can be submitted at a time. The roadmap includes further enhancements to streamline repository management and improve the overall contribution process.

GitHub Copilot has enhanced its context handling and model routing to maximize the utility of each session.

This improvement ensures that more of each session is dedicated to productive tasks, effectively extending the reach of user credits. By optimizing how context is managed and how models are routed, Copilot can allocate resources more efficiently.

A 30-game battle royale was conducted across eleven large language models (LLMs) to evaluate their performance, costing $482 in inference.

In this experiment, Grok 4.1 Fast emerged as the top performer, winning 43% of the matches, while Claude Sonnet 4.6 consistently attempted to collaborate with others. Interestingly, the least expensive model outperformed the priciest by 27 times in terms of cost per win. The findings highlight a discrepancy between typical benchmark results and practical applications of these models.

Z ai's GLM-5.2 has emerged as the leading open weights model on the Artificial Analysis Intelligence Index.

GLM-5.2 maintains the same size as GLM-5.

SpaceX is set to acquire Cursor for $60 billion.

The transaction highlights a significant investment in technology, with the deal expected to close in 2026. This acquisition marks one of the largest in the sector, underscoring the strategic importance of Cursor's offerings to SpaceX's future plans.

The article discusses the need for increased engineering discipline in the era of AI, contrasting it with previous shifts in infrastructure management.

The piece highlights how AI's rise recalls the transition from handcrafted server management to immutable infrastructure. The discussion touches on AI mandates, communication norms, and the ongoing debate over AI-generated code quality.