China vs US on AI

On June 13, 2026, at 5:21 p.m., Beijing and Manila time, China’s latest large language artificial intelligence model was released by Z.AI, formerly Zhipu, its flagship GLM 5.2, said to be the most powerful free language model in the world today.

The release spiked a 33 percent surge in share price of Zhipu (Knowledge Atlas Technology) from HK$950 to HK$1,486, on Monday, June 15, in the Hong Kong stock exchange. June 22, yesterday, Knowledge Atlas Technology was doing at HK$2,094, up 120 percent from HK$950, before the release of its GLM 5.2.

GLM 5.2 is said to match if not outperform Open AI’s two-month-old ChatGPT 5.5 in deep human-like thinking and multi-step reasoning and software engineering.

Per MindStudio: “GLM stands for General Language Model. GLM 5.2 is the latest generation of the GLM series – a large language model that supports text and multimodal inputs, long-context reasoning and coding tasks alongside its standout strength in design and creative visual understanding.”

Z.AI is a Beijing research company founded in 2019 as a spinoff from Tsinghua University’s Knowledge Engineering Group.

GLM excels in tasks like writing, translating or coding. It reads or matches patterns. A higher technology is Artificial General Intelligence (AGI) – a human-like super computer that can “reason, plan, learn and adapt across all human domains.”

Some analysts say the human race will see an AGI in five years, if not less.

When that happens, our Philippine Senate can disappear from the face of the earth and Filipinos will be better off for it. Because the intelligence of many of our senators poses little risk to the intelligence of even the most basic AI model. The crudest calculator can count beyond 24. Our senators cannot; they can count up to 12 – the legal quorum when one senator is in jail and another senator is hiding from the long arm of the law for previous acts of killing on a massive scale – humans, still the best computer models.

Wikipedia says “a Large Language Model (LLM) is an advanced type of artificial intelligence trained on massive amounts of text to understand, summarize, translate and generate human-like language. Powered by deep learning and transformer architectures, they operate essentially as advanced, large-scale probability engines that predict the next most likely word in a sentence.”

Why is there concern about GLMs and AGIs?

Because they impact on your daily life, lifestyle and security. Your security is being threatened. Everything you do and will do will be monitored by the government or a private company under contract with the government – as long as you carry a cellphone in your body or own a computer at home or in your office, or every time you are connected to the internet.

In China now, the government knows the whereabouts of up to one billion of its citizens, thanks to the ubiquitous cellular phone.

Government agents could kidnap you, load you into one of Elon Musk’s rockets and send you up in outer space or even to Mars, and no one would learn about it. Except AI. Most disturbing of all, nearly everyone, including possibly God Himself, risks losing his job. Because anything a human can do, an AI can do, and do better, including procreating. Machines can duplicate machines.

Why is the June 13 introduction of China’s GLM 5.2 such a stunning development in technology?

Because it means China has almost matched America’s advances in AI, at a much cheaper cost, faster pace and more efficiently. How? China simply copies the latest in AI that the United States develops and improves on it.

This is why President Donald Trump has banned foreigners or non-Americans from using Fable 5 of Anthropic, the rival of OpenAI. Anthropic and Open AI are America’s premier generative AI companies. OpenAI owns GPT 5. Anthropic owns Claude Opus, Sonnet and Haiku. OpenAI specializes in text, audio and image. Anthropic specializes in deep reasoning, long-context understanding, predictable responses, etc. Its AI models write better and know when you are asking them wrong questions deliberately.

Reports The Economist: “America’s lead over China in artificial intelligence may be at its smallest in over a year. When China disrupted the AI race in January 2025 with the release of DeepSeek R1, it erased $1T from America’s capital markets. Nvidia, a chip firm, briefly shed 17 percent of its value; the Nasdaq sank by 3.1 percent in a day. American investors were troubled not only because Chinese AI was good, but because it was being given away for free. The uproar soon faded. Since then market valuations around the world have hinged ever more on the promise that AI will be both revolutionary and profitable.” China matching America in AI was not supposed to happen – until 2030. This year, it seems it happened.

Now, says the magazine, “Chinese labs are unsettling their American rivals anew in the race to monopolize the market for models… GLM 5.2, (promises) ‘a step closer to frontier intelligence for everyone.’ It is the most capable Chinese-trained model to date and runs at less than a tenth of the price of Anthropic’s latest release, Fable 5. And as with other Chinese models, the weights, or parameters, that enable GLM 5.2 to function have been publicly released.”

Says The Economist: “Chinese models often excel in fields with clear right or wrong answers, like maths and coding. But they tend to fall down on problems that are open-ended or that require sustained independent judgement.”

“As the AI race speeds up, however, regulators everywhere will be faced with new challenges to safety and security. The risk of sudden government intervention may grow on both sides of the Pacific Ocean. Fable 5 was powerful enough to prompt such a response from the White House. That Chinese models are not, for now, subject to similar regulatory risk suggests China’s government is not yet alarmed enough to act.”

*      *      *

Email: [email protected]