"Combining software and hardware" breaks through the AI competition! Alphabet Inc. Class C (GOOGL.US) Cloud Next conference unveiled the "full set" of AI intelligence agents, with self-developed TPU 8t/8i making a simultaneous appearance.

In the increasingly fierce competition in the artificial intelligence (AI) field, Alphabet Inc. Class C (GOOGL.US) is attempting to regain its technological high ground through a strategy of combining software and hardware. At the Google Cloud Next conference held in Las Vegas this week, Alphabet Inc. Class C's cloud computing division not only introduced a new set of AI intelligence tools to compete with OpenAI and Anthropic, but also released the latest generation of its self-developed Tensor Processing Unit (TPU) series, challenging NVIDIA Corporation's dominance in the AI chip field. Software layer upgrade: AI intelligence agents take over the enterprise automation gateway At the annual conference in Las Vegas, Alphabet Inc. Class C's cloud division showcased tools that can be used to create AI intelligence agents and track their work processes within the enterprise, including a dedicated inbox for virtual intelligence agents to post information and progress reports. At the same time, Alphabet Inc. Class C also made several updates to its Workspace productivity suite and outlined a vision where AI intelligence agents will fundamentally change the daily work of ordinary employees. Many of the core technologies on which the current AI boom relies were created by researchers at Alphabet Inc. Class C. However, Alphabet Inc. Class C is now engaging in a fierce competition with leading AI intelligence agent manufacturers to win over enterprise customers wanting to improve productivity through AI technology. Alphabet Inc. Class C's capital expenditure for this year alone is expected to reach $185 billion, and investors are hoping that the company will develop enough new business to support this massive investment in AI. Google Cloud CEO Thomas Kurian stated in a blog post, "We are not providing individual services that can be pieced together, but building a complete underlying support system for innovation." The field of AI programming is a key area of focus for Alphabet Inc. Class C, and its management is increasingly concerned about falling behind competitors in this race. Several startup founders revealed that many engineers in Silicon Valley switch between Anthropic's Claude Code and OpenAI's Codex to compare their effectiveness, but rarely consider Alphabet Inc. Class C's related products. To attract developers, Alphabet Inc. Class C announced that its Gemini enterprise intelligence agent platform will add features such as a memory bank and personal memory archive to address the shortcomings of early AI tools that cannot retain historical interaction records. Another new feature, intelligence agent simulation, will help developers conduct more comprehensive testing before a tool goes live. Anthropic has extended its business reach to professionals in other industries through its Cowork product, and Alphabet Inc. Class C is also making efforts to compete in this market. Alphabet Inc. Class C introduced the Gemini enterprise application, positioning it as "the AI gateway for every employee," allowing users to create intelligence agents without writing any code. Alphabet Inc. Class C also launched the collaboration platform Projects, which supports employees and colleagues working together with AI intelligence agents. The platform can integrate various sources of information such as Workspace, Microsoft Corporation's OneDrive, and enterprise chat software to enable intelligence agents to operate within a complete context. Alphabet Inc. Class C also released other related products to help customers ensure the safe use of AI intelligence agents in industries with compliance requirements. In addition, Alphabet Inc. Class C also launched a new network security intelligence agent to help customers protect their systems. Although AI models can quickly identify a large number of vulnerabilities, the risk of them being maliciously exploited is also increasingly concerning in the absence of comprehensive protective measures. Hardware layer innovation: TPU 8t/8i debut, energy efficiency ratio soars over double At the infrastructure level of computing power, Alphabet Inc. Class C simultaneously introduced a new generation of TPU chip product line, attempting to further reduce the cost of AI inference and improve energy efficiency. At the conference, the company stated that the new TPU series will introduce two versions: TPU 8t for AI software development training, and TPU 8i designed specifically for the operational phase (or "inference" phase) of already developed AI services. In the current industry landscape dominated by NVIDIA Corporation, Alphabet Inc. Class C has become one of the most successful manufacturers of self-developed AI chips. In recent months, the demand for TPU chips in the Silicon Valley market has continued to rise, and Alphabet Inc. Class C hopes to sustain this momentum with its new generation of products. This new product release is a part of Alphabet Inc. Class C's overall strategy to lower the deployment costs and energy consumption of AI software, while also aiming to improve service response times. The new TPU chips are equipped with larger on-chip storage capacities, allowing for the rapid response expected by users. However, the increasing complexity of multi-layer software architectures is also leading to a continuous rise in computing power requirements. Mark Lohmeyer, Vice President of Cloud Computing and AI Infrastructure at Alphabet Inc. Class C, stated, "The core goal is to achieve as low a single transaction cost, and as low a response delay as possible. The total volume of transactions is growing significantly, and only by continuously reducing the cost of a single transaction can we support the scalable development of AI technology." The training TPU 8t chip can be deployed in clusters, and the system can be built into a super large-scale computing power system consisting of up to 9600 chips. Alphabet Inc. Class C stated that in deploying such large-scale computing clusters, power supply has become a major constraint for data centers, so operators need more energy-efficient systems to fully utilize limited power resources. The performance per watt of TPU 8t has increased by 124% compared to the previous generation, while TPU 8i has increased by 117%. The performance improvements are also attributed to the optimization of internal Network-1 Technologies, Inc. by Alphabet Inc. Class C, which greatly enhances the efficient communication capabilities between chips. Alphabet Inc. Class C's official statement announced that AI systems based on this chip series will be fully operational for commercial use later this year. Alphabet Inc. Class C also stated that it will continue to provide relevant services based on NVIDIA Corporation chips for customers in need, as NVIDIA Corporation products are currently the mainstream solution in the field of AI computing. Lohmeyer revealed that Alphabet Inc. Class C plans to be one of the first companies to deploy NVIDIA Corporation's new architecture hardware in the second half of this year.