A place to share my thoughts and reflections

Tumblr ↗

Tag: ai

  • The US’s AI Panic Isn’t About Theft. It’s About an Alternative Future.

    The Wall Street Journal recently told a compelling story: a former teacher and his pupil, once bound by academic dreams, are now the billionaire faces of China’s AI revolution, leaving US tech pioneers looking over their shoulders with alarm.

    Yet, within the first few paragraphs, the familiar accusations creep in. Theft. Espionage. National security. It is the reflexive mantra of Silicon Valley and Washington whenever China achieves a technological breakthrough.

    But here is the uncomfortable truth the West refuses to utter aloud: China hasn’t just caught up. It has built a fundamentally different vision for artificial intelligence—one that terrifies the US establishment not because it is stolen, but because it challenges their entire economic and social order.

    Let’s start with the allegation of theft. It is a psychological defense mechanism. It is far easier for US executives and politicians to accuse Chinese labs of copying than to admit a painful reality: US tech is bogged down by short-term greed. While US companies are busy figuring out how to charge you monthly for every chatbot query, Chinese engineers—under immense pressure from chip restrictions—were figuring out how to do more with less. The “student” surpassed the “master” through sheer grit, state-backed scale, and a homegrown talent pool that the US education system struggles to match. To concede this would force Wall Street to look inward and ask why their own innovation machine is stalling.

    But the deeper issue lies in ideology. The Western mind cannot process a non-capitalist tech boom. In the US playbook, any world-changing technology must ultimately serve the billionaire elite—it exists to enrich shareholders, consolidate monopolies, and extract recurring fees from the public. AI in the US is a luxury good, designed to replace workers and drive ad revenue for a handful of trillion-dollar corporations.

    China, on the other hand, treats AI like public infrastructure. Think of it less like a shiny new iPhone and more like high-speed rail or electricity. The goal isn’t just to create billionaires—though that happens—but to integrate smart systems into farming, disaster prediction and preparedness, logistics, and public healthcare. When China uses AI to forecast typhoons for coastal villages or optimize supply chains for its remote provinces, it doesn’t fit neatly into the Western narrative. The West dismisses this as propaganda because acknowledging it would raise a devastating question: Why is US AI only making the rich richer, while Chinese AI is being deployed to uplift the masses and eradicate poverty?

    This brings us to the “national security” boogeyman. Let’s be honest: banning Chinese AI products has very little to do with protecting citizens from spies. It is a trade barrier, plain and simple—a smokescreen for market protectionism. If US consumers and small businesses were allowed to use China’s highly efficient, cost-effective AI tools, the enormous profit margins of US tech giants would collapse overnight. Washington isn’t afraid that Chinese AI will steal your data; it is afraid that Chinese AI will steal the market by being better, cheaper, and more accessible. When you can’t out-compete, you outlaw.

    Finally, there is the ego problem. The West has long viewed itself as the eternal “teacher” of global technology. To watch the “pupil”—Tang Jie and Yang Zhilin—not only close the gap but innovate smarter under harsh restrictions, shatters that hierarchical arrogance. The US expected China to always be a few years behind. Now that the timeline has evaporated, the only face-saving move is to scream “cheating.”

    But the real threat Chinese AI poses to America is not espionage. It is the terrifying proof that there is another way to build the future. It is a future where technology serves the collective good, where progress is measured by societal resilience rather than quarterly earnings, and where the machine works for the many, not just the moguls.

    The West will not admit this narrative because doing so would unravel its ideological bedrock. If China can build advanced AI and use it for broad-based societal uplift, then America’s model—which creates immense wealth for a few while alienating the rest—is exposed as a choice, not an inevitability.

    So, they cry theft. They cry security. It is not an analysis of China’s capabilities; it is a confession of their own stagnation. And you cannot ban a better idea. You can only fear it.

  • Digital Sovereignty in the Age of AI: Why the Pasifika Must Send Its Best to Both Superpowers

    A quiet scandal sits inside Fiji’s immigration system. Every passport application, every visa, every piece of sensitive biometric data from Fijian citizens is collected through Jotforms—an American‑owned platform run by a Turkish national, hosted on foreign servers beyond our legal reach. Worse, just months ago, the system failed to open because it had “run out of paid or free quota.” A national border function, reduced to a cloud subscription.

    This is not sovereignty. This is digital negligence.

    Yet when someone points out that China’s open‑source AI revolution offers an alternative path—that models like DeepSeek are now neck‑and‑neck with America’s best, and freely available for any nation to download, inspect, and run locally—the response is sometimes dismissive: “Another DeepSeek preacher.”

    That dismissal misses the point entirely. Observing a strategic shift in global technology is not preaching. It is prudence. The real question is not whether to trust China or the United States. It is how a small Pasifika nation navigates between two superpowers without becoming a vassal of either.

    The Jotforms Lesson: Dependency Is Dangerous

    Our current reliance on Jotforms is a perfect example of what happens when no one is paying attention. We handed our citizens’ most sensitive data to a closed‑source, foreign‑controlled platform with no transparency, no local audit, and no fallback when the quota ran out. Where are our local IT geniuses? They are here—graduating from USP, from FNU, from our technical colleges. But they have never been given a strategic mission or the political backing to build national systems from scratch.

    The answer, however, is not to replace American dependency with Chinese dependency. That would be the same mistake, just a different flag. A Chinese cloud platform holding our passport data is no more sovereign than an American one.

    Non‑Alignment in the Age of AI

    For decades, Pasifika nations have practised geopolitical non‑alignment—engaging with Washington, Beijing, Canberra, and Tokyo on our own terms, refusing to be pawns in anyone’s cold war. We need the same doctrine for artificial intelligence.

    China’s open‑source AI models are a powerful tool precisely because they are open. We can download DeepSeek, run it on our own servers, modify it for our languages and needs, and never send a single citizen’s data across an undersea cable. That is fundamentally different from closed‑source American platforms like Jotforms, OpenAI, or Google Cloud, where the code is secret and the data leaves our jurisdiction.

    But the United States still leads in foundational research, chip design, and private investment. American universities like MIT—where young Penioni Narube from Cuvu Village is heading—remain world leaders. To ignore American AI would be as foolish as ignoring Chinese AI.

    A responsible Pasifika government sends its best to both.

    Three Concrete Steps

    1. Audit and reclaim. Conduct an immediate audit of every foreign digital service holding citizen data—Jotforms included. Migrate critical systems to open‑source, locally‑hosted alternatives using code from any nation, but always under our own physical and legal control.
    2. Send our best to both. Establish scholarships and exchange programs with both MIT and Tsinghua University, Stanford and Zhejiang University. Let our brightest software engineers learn AI from every major power, then return to build regional capacity.
    3. Declare digital non‑alignment. Adopt a formal policy that no exclusive cloud or AI agreement will be signed with any single country. Build interoperability, redundancy, and choice into every government system. Small nations survive by keeping options open.

    The Cost of Doing Nothing

    If we continue on our current path, we will wake up in five years to find our government data scattered across half a dozen foreign cloud platforms, our citizens’ faces and fingerprints held in servers we cannot inspect, and our AI future dictated by whichever superpower offers the cheapest subscription.

    That is not sovereignty. That is a quiet surrender.

    The open‑source AI revolution—accelerated by China’s DeepSeek—is not a reason to abandon caution. It is a reason to act. The tools are free. The talent is here. The only missing ingredient is the courage to take a shot, just as a young man from Cuvu did.

    Let us honour his journey by building a Pasifika that no longer needs to send its data abroad. Let us send our best to both superpowers—and then bring them home to build for ourselves.