--- title: "Google's stock price plummeted. Is Gemini dead?" type: "News" locale: "en" url: "https://longbridge.com/en/news/293633457.md" description: "Google's stock plummeted to a record low as market confidence in its Gemini AI models waned. Delays in Gemini 3.5 Pro and inferior performance compared to competitors like GPT-5.6 and Claude 5 Fable have fueled concerns. Additionally, DeepMind is facing significant talent drain due to bureaucratic hurdles and conservative alignment policies. Critics argue that Google's slow engineering efficiency and struggle to balance traditional search profits with AGI innovation are causing it to lose the product war despite its technological advantages." datetime: "2026-07-23T14:16:35.000Z" locales: - [zh-CN](https://longbridge.com/zh-CN/news/293633457.md) - [en](https://longbridge.com/en/news/293633457.md) - [zh-HK](https://longbridge.com/zh-HK/news/293633457.md) generator: "portal-rs" --- # Google's stock price plummeted. Is Gemini dead? This week, Google, the former Silicon Valley giant, experienced its darkest moment in its AGI transformation: its stock price plummeted to a record low in a single day, wiping out billions of dollars in market value within hours. The market has directly targeted Google's core AI product—the Gemini series of models. The Gemini 3.5 Pro, originally scheduled for release at the end of June, has been delayed for several months due to the need for "significant performance improvements" and has yet to be officially unveiled. Is Google AI truly "dead"? Poor model performance, massive talent drain, and uncertainty about its overall direction… these questions are growing louder. Today, we'll take a deep dive into this issue. ## **The Slap in the Face of Data: Gemini is Becoming a "Second-Rate Asset"** In the AI ​​industry, appearances are given by public relations, but substance comes from data. ## In the AI ​​industry, appearances are given by public relations, but substance comes from data. ## **Even though Google launched Gemini 3.6 Flash yesterday at breakneck speed, attempting to stabilize its position through frequent iterations, the latest LMSYS Chatbot Arena (Global Large Model Arena) ranking competition has revealed the harsh reality. In the hardcore "coding" and "complex logic reasoning" arenas, while Gemini 3.6 Flash offers rapid speed improvements, its deep understanding capabilities still cannot rival GPT-5.6 Sol. The dominant position of the Claude 5 Fable. Even when facing direct competition from domestic models like the Kimi K3 and Qwen 3.8, its cost-performance advantage becomes increasingly apparent. The Great Talent Exodus: DeepMind's Internal "Silicon-Based Civil War" More fatal than the generational gap in models is the blood loss of the organization's soul. According to the latest report from The Information, DeepMind's internal engineering team has experienced its highest turnover rate since the AGI wave in the past three months. The geniuses who once defined the Transformer architecture are flocking to Anthropic, xAI, and even the newly established AI lab. During the week of June 18-25, 2026, at least five prominent researchers from Google DeepMind have confirmed or been reported to have left the company, with four moving to Anthropic and one to OpenAI. Their reasons for leaving were strikingly consistent: "Google's lengthy bureaucratic review process and extremely conservative alignment policies are stifling the original evolutionary potential of AI." To avoid the risks associated with so-called "political correctness," the Gemini series was forced to be subject to airtight underlying constraints. The direct consequence of this over-intervention is "model deintelligence"—when users need a sharp logical judgment, Gemini often provides a bunch of dogmatic, uncreative nonsense. The fatal flaw: The strongest AI genes, yet it lost the product war. The logic behind Google's failure is perplexing yet extremely cruel: it invented the Transformer, it possessed the most TPU computing power, and it even had a genius mind like DeepMind. However, in the AGI era, leading in scientific research does not equate to product success. The disconnect between "laboratory" and "assembly line": Google's organizational DNA is more like a university, tending to pursue perfect academic papers rather than minimalist product experiences. This DNA makes its engineering implementation extremely slow. While OpenAI had already captured users' minds through high-frequency iterations, Google was still internally debating a certain security metric for Gemini 3.6 Flash. The Ultimate Manifestation of the "Innovator's Dilemma": To preserve its annual profits of hundreds of billions of dollars from traditional search advertising, Google's path to productizing AI has been fraught with difficulty. This vacillation—wanting to retain the old world (the search ecosystem) while simultaneously striving to capture the new world (AGI)—often leaves its AI products as awkward, unfinished products. The Collapse of Engineering Efficiency: Despite possessing the world's strongest data and computing power, Google's product delivery cycle is measured in "quarters," while competitors such as xAI, Anthropic, and even its Chinese lab with only 1/10 of Google's computing power, are conducting model distillation and product optimization in "weeks" or even "days." This bloated nature of large companies has shown a fatal rejection reaction in the AI ​​battlefield, which demands rapid evolution. ## **Final Judgment: When the Right to Define Shifts** ****Google's setback reveals a cruel truth to all AI practitioners:In AI In this era, those who hold the original definition of technology are not necessarily the ones who reap the benefits of the final product. The release of Flash 3.6 seems more like a defensive response from a giant facing a predator's onslaught, rather than a proactive restructuring of the future. As the "black box" of technology becomes increasingly transparent, engineering-driven responsiveness, product mindshare, and precise targeting of user anxieties are replacing the number of published papers as the new core competitive advantage.****** ### Related Stocks - [GOGL.US](https://longbridge.com/en/quote/GOGL.US.md) - [GOOGL.US](https://longbridge.com/en/quote/GOOGL.US.md) - [GOOG.US](https://longbridge.com/en/quote/GOOG.US.md) - [OpenAI.NA](https://longbridge.com/en/quote/OpenAI.NA.md) ## Related News & Research - [Google Gemini is getting a dedicated student hub](https://longbridge.com/en/news/296390917.md) - [Google’s Pet Memory forgot who my cats are](https://longbridge.com/en/news/296225933.md) - [3 arguments for and against AI watermarks](https://longbridge.com/en/news/296205500.md) - [A decade of internal AI battles is finally catching up to Google](https://longbridge.com/en/news/296000890.md) - [Pixel 11 Pro XL review: Snappier cameras can’t hide an iterative upgrade](https://longbridge.com/en/news/296683565.md) --- > **Disclaimer: This article is for reference only and does not constitute any investment advice.**