Google's Stock Falls Over 7% On Monday As Firm Faces Pressure, High-Profile Departures Alphabet Inc., the parent company of Google, saw its stock plummet 7.03% on Monday, trading at $342.16 on Nasdaq as of 11:17 am ET. The sharp decline followed a breakout week where the stock had previously rebounded from a five-week slide, gaining 2.3% before the downturn. The drop comes amid mounting pressures on the tech giant, including high-profile talent departures, legal and regulatory challenges, and investor concerns over its $80 billion equity raise to fund AI initiatives. A key factor in the market reaction was the announcement of a "first-of-its-kind partnership" between Google DeepMind and A24, a prominent film production company. The collaboration, detailed in a company release, aims to advance research in AI-driven creative processes, enabling artists to explore new techniques and workflows. However, this partnership did not alleviate investor anxieties, as other issues continued to weigh on the stock. The company’s legal troubles intensified when a California court denied motions by Google’s YouTube and Meta for a new trial after a jury found both firms liable for designing social media platforms that harm young users. The verdict imposed $6 million in damages on the tech giants, highlighting growing regulatory scrutiny over their impact on mental health and societal well-being. High-profile departures from Google’s AI division further fueled investor concerns. Nobel laureate John Jumper, a co-recipient of the 2023 chemistry prize, announced his departure from DeepMind to join rival Anthropic. Jumper was a pivotal figure in the development of Google’s coding tools and had played a central role in advancing AI research within the company.#youtube #meta #a24 #alphabet_inc #google_deepmind
Google Loses Two Top AI Researchers To OpenAI & Anthropic Noam Shazeer, a co-lead of Google’s Gemini models, and John Jumper, the architect of the AlphaFold protein prediction system, have both announced their departures from Google DeepMind. Shazeer joined OpenAI, while Jumper is moving to Anthropic, marking a significant shift in the competitive landscape of AI research. Shazeer’s departure was announced on June 18, with his new role at OpenAI confirmed by CEO Sam Altman. Jumper, who won the 2024 Nobel Prize in Chemistry for his work on AlphaFold, plans to take a break before starting at Anthropic, where his expertise in protein structure prediction aligns with the company’s focus on AI for scientific applications. Shazeer’s contributions to AI include co-authoring the seminal “Attention Is All You Need” paper, which introduced the Transformer architecture now foundational to large language models. Google had previously brought him back in 2024 through a reported $2.7 billion deal with Character.AI, positioning him as a co-lead on the Gemini project. However, he left less than two years later, raising questions about Google’s ability to retain top talent. Jumper’s work on AlphaFold revolutionized protein structure prediction, a breakthrough that has since become a cornerstone of biological research. Both researchers confirmed their moves, with Google DeepMind and Anthropic acknowledging the departures. The timing of the exits coincided with a sharp decline in Alphabet’s stock, which fell approximately 5% to 6% on June 22. Market analysts linked the drop to concerns over Google’s AI spending and its capacity to retain senior researchers. The stock had previously stabilized after initial reports of Shazeer’s departure, but the broader implications of losing two high-profile figures weighed on investor confidence.#sundar_pichai #anthropic #openai #sam_altman #google_deepmind
Google Unveils Gemini 3.5: A Leap in AI Capabilities for Agents and Coding Google DeepMind has launched Gemini 3.5, a new family of AI models designed to enhance complex workflows and coding tasks. The release includes 3.5 Flash, a model optimized for speed and performance, and 3.5 Pro, which is already being tested internally and will be rolled out next month. The announcement highlights advancements in agentic intelligence, multimodal understanding, and real-world applications across industries. Gemini 3.5 Flash is now available globally through the Gemini app, Google Search’s AI Mode, and developer platforms like Google Antigravity and the Gemini API. It is positioned as a powerful tool for developers and enterprises, offering capabilities that rival large flagship models in performance while maintaining exceptional speed. The model excels in coding, long-horizon tasks, and multimodal reasoning, achieving notable benchmarks such as 76.2% on Terminal-Bench 2.1, 1656 Elo on GDPval-AA, and 83.6% on MCP Atlas. Its output token speed is four times faster than other frontier models, balancing quality and latency. The release emphasizes the model’s ability to handle complex, real-world problems. For instance, 3.5 Flash can automate tasks like renaming and categorizing unstructured assets, synthesizing academic papers into playable games, and transforming legacy codebases into modern frameworks like Next.js. It also supports collaborative subagents through the Antigravity platform, enabling scalable solutions for tasks such as financial document preparation, data analysis, and application development. Industry partners are already leveraging Gemini 3.5 Flash for transformative applications.#google #google_deepmind #shopify #gemini_35 #google_antigravity
Google's New Gemma 4 Models Bring Complex Reasoning Skills to Low-Power Devices Google LLC has launched its latest open-weight artificial intelligence models, Gemma 4, marking a significant advancement in the field of lightweight, high-performance AI. These models, built on the architectural foundation of Gemini 3, are designed to handle complex reasoning tasks and support autonomous AI agents running on low-power devices such as workstations and smartphones. The release positions Google as a key player in the growing market for edge computing and local AI applications. The Gemma 4 family includes four variants: Effective 2B, Effective 4B, a 26B Mixture of Experts (MoE) model, and a 31B Dense model. The smaller "Effective" models are tailored for edge use cases, such as Android smartphones and Raspberry Pi computers, while the 26B MoE model introduces an innovative approach by activating only 3.8 billion parameters during inference tasks. This optimization allows the model to maintain high performance without compromising the depth of knowledge typical of larger models. The 31B Dense variant currently ranks third in open models on the industry-standard Arena AI Text leaderboard, demonstrating its competitive edge. Google DeepMind researchers Clement Farabet and Olivier Lacombe highlighted the models' ability to deliver "more intelligence per parameter," enabling them to outperform their size class. This efficiency is critical for applications requiring real-time processing and minimal computational resources. The models are also engineered to support AI agents, with native capabilities for function calling and structured JavaScript Object Notation (JSON) outputs.#google_llc #google_deepmind #clement_farabet #olivier_lacombe #hugging_face
