JD.com is developing China's AI industrialization plan, opening up the physical AI stack from EgoLive data to robotics bases

Automated chip design using artificial intelligence has become one of the most discussed topics in the industry. Last month, the Chinese company Kimi demonstrated a complete experiment where its K3 model independently designed a chip. This chip was tested on an EDA platform, running for 48 hours exclusively on AI, without using commercial software from Cadence or Synopsys.

However, a deeper analysis shows that replacing existing systems is still far from complete: the chip from Kimi corresponds to technologies from about 20 years ago, runs 20–30 times slower than modern chips, and open-source EDA tools still require commercial systems to perform real projects.

Nevertheless, agentic AI is truly transforming workflows, covering tasks from generating RTL code to creating test environments and invoking simulation, formal verification, and debugging tools.

At the DAC Chips to Systems 2026 conference, three leading EDA companies presented their different strategies. Synopsys, possessing the largest accumulation of AI, introduced a fully autonomous agent for design verification and an autonomous thermal simulation workflow based on Ansys Icepak. Furthermore, the company showcased an analog/mixed-signal workflow with a 3x efficiency increase and 20 GPU-accelerated tools. Synopsys proposed an AI autonomy ladder in the style of self-driving transport, from L1 to L5, which is currently at level L3, and launched AgentEngineer in 2026 after DSO.ai in 2020. In November 2025, NVIDIA invested two billion dollars in Synopsys, and Grace Blackwell is expected to provide 30 times higher performance for EDA workloads.

Cadence's strategy focuses on architectural completeness. The company introduced AuraStack—an AI-based superagent that joins ChipStack, InnoStack, and ViraStack to ensure coverage for verification, digital implementation, analog design, as well as printed circuit boards and packaging. This stack provides 20 times higher multiphysics analysis performance and a 15x workflow acceleration. NVIDIA engineers are already using this system. Cadence's most distinctive innovation is the Mental Model—a structured representation of knowledge that fixes design intentions to reduce hallucination risk, and its agents have been recognized by NVIDIA, Qualcomm, and Broadcom.

Siemens EDA adheres to a different approach—trusted autonomy. The core logic of their Fuse EDA AI agent uses self-checking, matching the results of large models with deterministic, physics-based assertion engines. The Solido Layout Analyzer tool helped the STMicroelectronics non-volatile memory team reduce debugging time by weeks, and through Intelligence Center X, Siemens is expanding agent planning to production processes and supply chains. Fuse uses the MCP protocol to connect tools from different vendors. All three giants rely on NVIDIA inference level, management level, and the Vera processor in EDA verification farms.

Chinese manufacturers view the current situation as a favorable moment. Xpeedic, together with Lenovo, presented its EDA agent as the only example of domestic EDA at DAC 2026. XEPIC launched the basic Agentic EDA XEPIC Intelligent System with intelligent verification and design, focusing on proof cycles that transform AI outputs into verifiable, traceable results ready for approval. UniVista released UDA 2.0—the first agentic EDA tool on a fully proprietary domestic architecture. Other solutions include Chips Z for advanced packaging, DVcrew and PDcrew agents from Banxin, and an LLM-based verification platform from Zhiwei Chuangxin called ChatDV, which reduces modular verification from months to hours. NVIDIA research shows that chip designers spend about 60% of their time on debugging—precisely the area targeted by agentic AI.

The Chinese artificial intelligence industry has experienced a renewed momentum in recent weeks, driven by a series of releases that have narrowed the gap with major US developers. Companies such as Alibaba, Moonshot AI, DeepSeek, Z.ai, and ByteDance are now competing among the most sophisticated available models.

This movement began earlier this year when Chinese companies started presenting systems capable of handling complex tasks, including programming, reasoning, and video creation. The adopted tactic combined technical improvements with lower prices compared to those practiced by American competitors.

This expansion of models has intensified pressure on giants like OpenAI and Anthropic, forcing them to face a scenario where both high performance and cost efficiency have become crucial for attracting users and developers globally.

The new phase of the AI efficiency battle

This recent wave from China signals a transformation regarding the initial impacts caused by the advancement of DeepSeek. The sector's progress is no longer seen as an isolated event but is interpreted as the result of an ecosystem capable of generating successive competitive models in short periods.

Notable releases include Qwen3.8-Max, developed by Alibaba and presented as the company's most advanced model; Kimi K3, from Moonshot AI; V4 Flash, from DeepSeek; GLM-5.2, from Z.ai; and new iterations of ByteDance's Seedance tool, focused on AI video generation.

The main focus of the rivalry has shifted from simply developing the most powerful system to prioritizing cost reduction. Independent evaluations mentioned in the text indicate that certain functions performed by DeepSeek V4 Flash could cost mere cents, while comparable American solutions would reach significantly higher values.

This landscape has put pressure on intermediate market companies. Companies that cannot offer greater capacity at the same price or reduce their costs risk losing market share, as users have begun to directly compare the value offered versus the performance.

The competition has also changed how some corporations utilize various models. According to Dermot McGrath, founder of the consultancy ZenGen Labs, some professionals have opted to integrate American and Chinese tools: using US models for planning and Chinese systems for executing specific tasks.

The Chinese advance carries both a technological and political dimension. Washington has expressed concern over issues such as intellectual property and the effect of these developments on its sectoral leadership, while simultaneously discussing methods of control without harming its competitive position.

Another factor highlighted in Chinese growth is the willingness of companies to accept lower financial returns in the short term with the goal of capturing users, developers, and international relevance. This strategy has fueled an intense price war within the Chinese market itself.

Beyond language models, the video creation area exemplifies this expansion. ByteDance has increased its investments in Seedance, while Kuaishou has progressed with its Kling AI tool, which received financial support from Alibaba and Tencent in a $2.8 billion round.

At the heart of this race lies the dispute over technological supremacy. While American laboratories maintain plans for large market valuations based on high-margin business models, Chinese competition questions to what extent these prices can remain high against more accessible alternatives.