New GPT-5.4 clobbers humans on pro-level work in OpenAI's tests - by 83%

GPT-5.4 is also more reliable, producing 18% fewer errors and 33% fewer false claims than GPT-5.2, according to OpenAI.
eWeek

Google Launches Canvas in AI Mode for Users Across the US

Google rolls out Canvas in AI Mode across the US, adding a side panel workspace in Search for building plans, dashboards, and interactive tools.

Google’s AI Mode in Search now lets you plan, write, and code on its own Canvas

Google Search has updated Canvas inside AI Mode, a workspace powered by Gemini that already lets users draft and refine documents, to now support coding projects and interactive tools. With the latest ...
IEEE

Algorithmic Verification of Modular Multiplicative Divisor Labeling in Vertex Switched Jewel Graphs with Odd/Even Jewel Count for Secure Networks Using Python Pseudocode

Abstract: This research investigates the characteristics of Modular Multiplicative Divisor MMD labeling in a Jewel Graph under vertex switching when the jewel count is odd. We formally establish that ...
eWeek

6 Best Nano Banana 2 Prompts in 2026: How to Generate Smarter AI Images

Discover the best Nano Banana 2 prompts to test Gemini 3.1 Flash Image, from 4K mockups to multilingual text and character consistency.