Zhipu CEO's Internal Letter: Beyond the Trillion-Dollar Valuation, Focus Shifts to AGI
Zhipu founder and CEO Tang Jie issued an internal letter on July 11, 2026, titled "The Giant Wave is Coming," reflecting on the company's recent successes and outlining future strategic directions. Over the past six months, Zhipu has experienced significant growth, with its market value increasing tenfold since its IPO and joining the HK$1 trillion club in June 2026, surpassing major tech companies like Baidu and Xiaomi. This success is attributed to a strategic bet on AI coding capabilities made in early 2025, which has proven to be a critical differentiator in the competitive AI landscape. The company's flagship models, GLM-4.5 and the open-sourced GLM-5.2, have positioned Zhipu among the global AI coding leaders, with GLM-5.2 matching or exceeding benchmarks set by competitors like Anthropic's Claude Opus 4.8 and OpenAI's GPT-5.5. Zhipu's financial performance also reflects this growth, with its MaaS platform achieving an ARR of 1.7 billion yuan by March 2026, a 60-fold increase over the previous year. Looking ahead, Tang Jie announced Zhipu's "Touch High" plan, a strategic initiative focused on pushing the boundaries of artificial general intelligence (AGI) rather than immediate commercialization. This plan will concentrate on four core areas: long-horizon task capabilities, fully autonomous agent systems, self-evolving AI, and robust safety and governance mechanisms. The company is committed to open-sourcing its advanced models, like the recently released GLM-5.2, to foster an inclusive AI ecosystem and ensure widespread access to cutting-edge technology.
Zhipu's strategic pivot from rapid commercialization to a long-term AGI focus, as articulated by CEO Tang Jie, highlights a critical inflection point in the AI industry. The company's valuation surge and market positioning underscore the significant returns from its prior investment in coding capabilities, a testament to the power of anticipating technological shifts. However, the "Touch High" plan's emphasis on AGI, long-horizon tasks, and autonomous agents signals a move towards more fundamental, potentially disruptive research. This approach, while ambitious and aligned with the industry's ultimate goals, carries inherent risks, particularly concerning resource allocation and the uncertainty of achieving AGI within a defined timeframe. The commitment to open-sourcing advanced models like GLM-5.2 alongside this ambitious research agenda presents a dual strategy: driving innovation while attempting to democratize access and foster a collaborative ecosystem. The critical challenge will be balancing the immense computational and research demands of AGI pursuit with the need for sustainable commercial viability and robust safety protocols, especially as AI capabilities approach human-level or superintelligent thresholds.
AI-generated to prompt reflection — not editorial opinion, not advice, not a statement of fact. How this works.