GitHub Copilot

GitHub Copilot now supports a 1M-token context window and configurable reasoning levels in VS Code, Copilot CLI, and the Copilot app. It matters because larger repo context and explicit effort controls arrive in the same surfaces teams already use for agent-assisted development.

Microsoft introduced MAI-Thinking-1, MAI-Code-1-Flash, and five other MAI models across code, image, voice, and speech. The launch puts Microsoft back into the frontier-model race and starts landing pieces of the stack in Copilot and partner runtimes.

GitHub introduced a desktop Copilot app as a new home for agent-native development, with canvases and continuity across local, cloud, mobile, and web workflows. It matters because Copilot is moving from inline assistance toward a multi-surface agent workspace for longer-running development tasks.

GitHub made remote control generally available for Copilot CLI and code sessions, so users can monitor runs, approve actions, and answer prompts remotely. That turns long-running coding jobs into asynchronous workflows instead of terminal-bound sessions.

Ahead of GitHub Copilot's June 1 usage-based billing switch, users documented GPT-5.5 sessions hitting 60M tokens and $221 across 15 messages on the legacy per-message plan. The examples show why flat message buckets break once single requests can run for hours and consume extreme token counts.

VS Code briefly enabled AI co-author tagging by default in its Git extension, then a maintainer said the default will revert to off in 1.119. If you rely on clean commit output, check your Git settings before upgrading.

GitHub expanded semantic indexing beyond GitHub and Azure DevOps remotes, so Copilot can search across more workspace types and repositories inside @code. That improves agent context retrieval in local workflows, while the same release also adds chat-history recall and prompt-eval tooling.

GitHub says Copilot will shift from flat-rate plans to usage-based billing starting June 1 as agentic features expand. The change makes token budgeting a first-order engineering constraint and adds more pressure on teams comparing Copilot with other coding agents.

GitHub added bring-your-own-model keys to Copilot in VS Code, letting users connect local or cloud providers instead of only bundled models. Teams can keep the Copilot harness while routing prompts through approved backends such as LM Studio or OpenRouter.

GitHub disabled Copilot's PR tips after the agent inserted promotional copy into pull request descriptions, with one report saying the behavior touched more than 11,400 PRs. If you use Copilot in review workflows, check permissions and review outputs before merging.

GitHub said Copilot Free, Pro, and Pro+ interaction data will train models by default from Apr. 24 unless users opt out, while private repo content at rest stays excluded. Teams should review per-user enforcement, enterprise coverage, and repo privacy settings before the change lands.

GitHub will start using Copilot interaction data from Free, Pro, and Pro+ tiers for model training unless users opt out, while Business and Enterprise remain excluded. Engineers should recheck privacy settings and keep personal and company repository usage boundaries explicit.

Microsoft introduced Copilot Cowork for Microsoft 365, turning requests into plans that execute across apps and files within enterprise security and governance boundaries. Teams evaluating office agents should watch approvals, app coverage, and durable execution across its multi-model setup.