Full Summary
This Friday, July 17th, two major AI developments are dominating the news: Google's Gemini 3.5 Pro is delayed, while China's Moonshot AI has launched Kimi K3, a model many sources confirm is challenging US AI dominance. Both PYMNTS.com and India Today report Google's Gemini 3.5 Pro faces a significant delay because its coding capabilities need a boost. The Star and WION add that the model failed to meet internal goals, causing frustration among Google employees and leading to a nearly 3% drop in Alphabet shares. Google CEO Sundar Pichai announced Gemini 3.5 Pro in May, but engineers are reportedly unhappy with its performance, particularly in coding. But then, an even bigger story: Moonshot AI, a Chinese startup, has launched Kimi K3, which SiliconANGLE, Fortune, and Neowin all confirm is the world's largest open-weight AI model, boasting 2.8 trillion parameters. Multiple sources, including CNBC and AI Magazine, state Kimi K3 outperforms OpenAI’s GPT 5.5 and Anthropic’s Opus 4.8 across most benchmarks, especially in coding. Business Insider and AOL.com highlight that Kimi K3 even topped Arena.ai's front-end coding leaderboard, surpassing Fable 5 and GPT 5.6 Sol. Fortune further reports that Moonshot AI is pricing K3 at a fraction of the cost of its US rivals, at $15 per million output tokens compared to Fable 5’s $50. Moonshot AI plans to release the full model weights for Kimi K3 by July 27th, making it accessible for organizations to customize and run. This development, confirmed by International Business Times and Taipei Times, signals a significant shift in the global AI race. This means that while US tech giants face internal setbacks, Chinese AI is rapidly advancing, potentially offering powerful, cheaper alternatives that could impact your access to cutting-edge AI tools and even shift investment strategies in the global tech market.