Talent drain, model lagging, commercial pressure! The user count of Gemini surpassing 1 billion cannot mask the "internal and external troubles" facing Alphabet Inc. Class C (GOOGL.US).

As competition continues to intensify among leading global chat providers like Siasun Robot & Automation, Sundar Pichai, CEO of Alphabet Inc. Class C (GOOGL.US), revealed that the user count for the company's Gemini application has now surpassed one billion. Pichai posted on social media on Tuesday, stating, "Currently, more than one billion people are using the Gemini application each month to spark new ideas and get work done. This is our fastest-growing product to date and our 14th product to reach the milestone of one billion users." Pichai added, "Kudos to Josh Woodward, Inc. and the entire Gemini team, and thanks to everyone who has been with us on this journey theres even more exciting content to come!" Woodward, Inc. is the Vice President of Alphabet Inc. Class C, responsible for leading Google Labs, the Gemini application, and AI Studio. Continued Competitive Pressure The milestone of one billion users for the Gemini application is undoubtedly exciting news for Alphabet Inc. Class C, but the company still faces multiple challenges in the artificial intelligence (AI) sector. The first comes from external pressures. OpenAI announced on July 31 that its AI model currently reaches over one billion active users, with two million enterprise users. To enhance its commercialization process, OpenAI has expanded its ChatGPT advertising business. On Tuesday, the company led by Sam Altman disclosed that it has launched ChatGPT Ads in the UK, Mexico, Brazil, Japan, and South Korea. OpenAI began testing ChatGPT ads in the United States in February, and by March, the service had expanded to Canada, Australia, and New Zealand. OpenAI stated, Maintaining the fast and reliable operation of the Free and Go plans requires significant infrastructure and ongoing investment. Advertising helps fund this work by providing higher-quality free and low-cost options to support broader AI use and allows us to continuously enhance the intelligence levels and capabilities we offer. If you prefer not to see ads, you can upgrade to the Plus or Pro plans or choose to turn off ads in the Free plan, but correspondingly the number of free messages available each day will decrease. OpenAI has indicated that advertising will not affect the output results of ChatGPT. Despite SpaceX AI's Grok significantly trailing behind Gemini, ChatGPT, and Anthropic's Claude in market share, it still boasts a loyal user base and continues to roll out regular updates and new features. On Tuesday, SpaceX AI launched Grok Bot, which functions similarly to a team of always-on intelligent agents that can work across different applications and tools. SpaceX AI stated, We initially built Grok Bot as an internal prototype, and it quickly grew throughout the company. Various teams create Bots to handle sales outreach, marketing campaigns, office operations, bug fixes, and other tasks. Today, we are excited to make it available to users worldwide. The product has been made available in beta form starting that day, targeting users subscribed to SuperGrok Heavy, Cursor Ultra, and Cursor Teams Premium on Apple's iOS system. The company stated, Employees at SpaceX AI commonly run multiple Bots simultaneously, with one managing the others. The top level features a Chief of Staff, with a specialist agent assigned to each work area inbox management, expense management, recruitment, bug fixes, or operations. You no longer need to manage multiple intelligent agents; Grok Bot provides you with a lean team that can work in parallel, so you won't need to serve as the middleman. Alphabet Inc. Class C May Have Shelved Release of Gemini 3.5 Pro AI Model In addition to external challenges, signals have emerged from within Alphabet Inc. Class C that have left some market observers uneasy. According to a recent report by industry analysis firm Semi Analysis, it appears that Alphabet Inc. Class C may have shelved the Gemini 3.5 Pro model. During the I/O Developer Conference held in May of this year, Pichai indicated that the Gemini 3.5 Pro model would be launched in June, but it has yet to go live. Originally, the Gemini 3.5 Pro model was expected to carry the responsibility of re-establishing Alphabet Inc. Class Cs stake in AI frontier competition. After all, in a context where OpenAI and Anthropic continually refresh model capabilities, the Gemini Pro series has been a key benchmark for evaluating Alphabet Inc. Class C's AI strength. There is speculation that the actual capabilities of the Gemini 3.5 Pro model may not have met Alphabet Inc. Class C's initial expectations. Semi Analysis believes that the capabilities of Gemini 3.5 Pro are roughly on par with Anthropic's Claude Opus 4.5, which is already a model released in late November last year. As a transitional solution, Alphabet Inc. Class C has rolled out more Flash models, including 3.6 Flash, 3.5 Flash-Lite, and 3.5 Flash Cyber. However, current public evaluations suggest that the Gemini Flash series emphasizes speed, cost, and deployment efficiency, but still lags behind Muse Spark 1.2, Grok 4.5, and leading open-source models in complex reasoning, programming, and agent tasks. During the second-quarter earnings call for Alphabet Inc. Class C in 2026, Pichai reiterated, I am very excited about our progress in the frontier space, and expressed confidence in maintaining the leading position of the Gemini models. However, Semi Analysis pointed out that Gemini has clearly fallen behind frontier models. According to the Artificial Intelligence Index, Gemini currently ranks 8th in the lab rankings, behind Muse Spark 1.2, Grok 4.5, and top models from Chinese open-source labs. Talent Exodus Earlier this month, Chief Scientist Jeff Dean announced his departure after 27 years. Prior to that, several notable researchers had already left Alphabet Inc. Class C, including Noam Shazeer, one of the authors of the milestone 2017 paper Attention Is All You Need, which laid the foundation for generative AI; all eight authors of that paper have now left Alphabet Inc. Class C. Shazeer joined OpenAI in June this year, less than two years after Alphabet Inc. Class C spent nearly $3 billion to bring him back through acquihiring. Shortly after his departure, Nobel laureate John Hennessy also left DeepMind for Anthropic. Analyst Gil Luria from D.A. Davidson pointed out that the trend of top talent leaving Alphabet Inc. Class C is evident. He stated, They are not eager to commercialize AI; rather, they want to be part of history. Therefore, they view Anthropic, OpenAI, or other startups as places where they can write history. Alphabet Inc. Class C's investments in global data centers, chips, and related infrastructure are unmatched, yet computing power remains in short supply. Every TPU allocated for training models, supporting Alphabet Inc. Class C products, or fulfilling cloud customer contracts represents a decision among multiple priorities. According to several anonymous sources, some researchers within Alphabet Inc. Class C are increasingly dissatisfied with their ability to secure computing powerthey find it challenging to obtain the computational resources needed to advance frontier projects while seeing Alphabet Inc. Class C sell its in-house TPUs to external clients, including Anthropic. Additionally, the hierarchical approval process within Alphabet Inc. Class C is cumbersome, and turning research results into products requires multiple levels of clearance, making companies like OpenAI, Anthropic, and younger startups more attractive to AI researchers they prefer lab work over quarterly financial results. AI Leadership Overhaul Recent reports indicate that there has been a substantial adjustment in the AI leadership at Alphabet Inc. Class C. DeepMind co-founder Demis Hassabis has relinquished day-to-day management responsibilities and stepped down as the overall head of Gemini commercialization, taking on a role as Chairman of DeepMind and Chief Scientist of Alphabet Inc. Class C, focusing mainly on long-term AI research. Former DeepMind Chief Technology Officer Koray Kavukcuoglu has been promoted to Senior Vice President at DeepMind (DeepMind will no longer have a separate CEO), responsible for the development and operation of the Gemini model and reporting directly to Pichai. In addition, Alphabet Inc. Class C co-founder Sergey Brin has become more directly involved in Gemini. Currently, Brin has no formal executive position; however, since the release of ChatGPT, he has re-engaged with Alphabet Inc. Class C's daily AI affairs, participating in model testing, discussing technical direction, and exerting increasing influence over the development direction of Gemini. The core of this restructuring is to move the AI decision-making center back from London to Silicon Valley, aiming to reverse a trend that has plagued the company since at least 2023. That year, Alphabet Inc. Class C merged two previously independent and highly valued scientific labs: one located at the company headquarters in Mountain View, the other rooted in London at DeepMind. Despite the merger being under the name of DeepMind, researchers remained divided across continents. Sources indicate that this arrangement complicated the decision-making process and left talent in both regions feeling dissatisfied. A major focus of Alphabet Inc. Class C's future business strategy is to transform the "research federation" previously scattered between Londons DeepMind and California's Google Brain legacy framework into a Gemini product delivery machine centered in Mountain View, reporting directly to Pichai. Kavukcuoglus dual role overseeing both the day-to-day operations of Google DeepMind and serving as Chief AI Architect means that model pre-training, post-training, evaluation, computational resources scheduling, and collaboration with Cloud, Search, and developer products will fall under a shorter decision chain. These latest personnel adjustments are not intended to weaken foundational research but rather to manage long-term scientific exploration and quarterly product delivery separately, attempting to address the slow decision-making and commercialization of research outcomes that have plagued cross-continental teams since the 2023 merger. The Gemini 3 series models and the Nano Banana image editing tool have helped Alphabet Inc. Class C regain market attention. The company has also proven that it still possesses the capability to develop cutting-edge models. However, the more challenging task now is to rapidly translate model capabilities into developer tools, enterprise services, and stable revenue. In the AI programming and enterprise market, OpenAI and Anthropic have already gained a first-mover advantage. Alphabet Inc. Class C's executives and board are concerned that the company's research strength has not yet fully translated into product competitiveness. Previously, Gemini encompassed multiple functionalities, including research, modeling, applications, and security. The lengthy decision chain has made it easy for research goals and product timelines to conflict. The new management structure delegates daily development of Gemini to Kavukcuoglu, helping to shorten the decision cycle and accelerate model releases and product implementations. Reports also mentioned that some Alphabet Inc. Class C executives expressed dissatisfaction with Hassabiss level of commitment to commercialization work. AlphaFold has been viewed as a controversial case. This protein structure prediction system helped Hassabis secure the 2024 Nobel Prize in Chemistry and brought tremendous impact to scientific research; however, after the project was opened up for free, Alphabet Inc. Class C did not achieve commercial returns commensurate with its investment scale. People close to Alphabet Inc. Class C denied any significant conflicts between the two parties. They pointed out that Hassabis led the early release of Gemini and has consistently cared about project progress. Following significant shifts in Alphabet Inc. Class C's AI leadership, whether the new structure can prove effective depends on the smooth collaboration of three routes. Research needs to retain exploratory space, model teams must increase iteration speed, and the product division must find willing paying users. For Alphabet Inc. Class C, top papers and model rankings are no longer sufficient to alleviate external doubts; the upcoming competition will hinge on code tools, enterprise markets, product experiences, and commercial revenue. Brins return to the AI power center signifies that Alphabet Inc. Class C regards Gemini as a core battle requiring the personal involvement of its founders. Hassabis stepping back into a front-line research role also indicates that the research-led model established at DeepMind for over a decade is undergoing change. Alphabet Inc. Class C possesses models, computing power, data, distribution channels, and a vast enterprise customer base. Now, it needs to demonstrate its ability to compress these resources into a faster product pipeline.