Google I/O Blitz: Gemini 3.5, Omni, and a GitHub Nightmare
In today's episode:
* Google rolls out Gemini 3.5 Flash, its “strongest agentic and coding model yet”, for tackling long-horizon agentic tasks, in the Gemini app and Search's AI Mode [https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-5/]
* Gemini 3.5 Flash costs $1.50 per 1M input tokens and $9 per 1M output tokens, 3x the price of Gemini 3 Flash Preview and 6x the price of Gemini 3.1 Flash-Lite [https://simonwillison.net/2026/May/19/gemini-35-flash/]
* GitHub confirms ~3,800 internal repositories were breached after one of its employees installed a malicious VS Code extension; TeamPCP claims responsibility [https://www.bleepingcomputer.com/news/security/github-confirms-breach-of-3-800-repos-via-malicious-vscode-extension/]
* Google launches Gemini Omni, a multimodal model it says can “create anything from any input”, starting with video generation, for Google AI Plus, Pro, and Ultra [https://venturebeat.com/ai/google-unveils-gemini-omni-any-to-any-ai-model-what-enterprises-should-know]
* Google updates Flow and Flow Music with Gemini Omni Flash support, mobile apps, the ability to create custom tools like a video resizer or shaders, and more [https://www.cnet.com/tech/services-and-software/google-doubles-down-on-ai-content-creation-with-updates-coming-to-flow-and-flow-music/]
* Google adds a conversational search feature to YouTube, initially for Premium users, and integrates Gemini Omni into YouTube Shorts Remix and the Create app [https://www.androidcentral.com/apps-software/youtube/google-i-o-youtube-updates]