Miin

30-second overview: Miin is a news aggregation platform released by Taiwan AI Labs, developed under the leadership of PTT Bulletin Board System founder Ethan Tu (杜奕瑾). It uses AI to automatically organize coverage of the same event from different media outlets, labels stance differences, and detects "coordinated operation accounts" on social platforms. During House Speaker Pelosi's visit to Taiwan in 2022, Miin's analysis showed that one in every four Twitter accounts related to the event was involved in cognitive operations.¹ But at the end of 2025, NEXTAPPLE News accused Miin's AI crawler of scraping more than 250 exclusive news articles without authorization, and Tu was reported to the authorities for suspected copyright violation.² Miin's story is a frontline snapshot of AI fighting disinformation colliding with news industry copyright.

In 1995, Ethan Tu built PTT in his National Taiwan University men's dormitory on a 486 computer.³ Thirty years later, the problem he faces is no longer "how to let people discuss freely," but "how to help people discern truth from falsehood in a manipulated information environment."

"If social platforms maximize user engagement and polarize users, the platform maximizes advertising revenue." That is how Tu analyzed it in an interview with CommonWealth Magazine.⁴ Miin's design logic is the inverse: no algorithmic push, no pursuit of engagement — instead, it presents side-by-side coverage of the same event from different media outlets, letting readers judge for themselves.

The AI That Detects Water Armies

Miin's most core technology is not news aggregation — it is "coordinated behavior detection."

Traditional fact-checking is a lagging indicator: by the time the fact-check result is out, disinformation has already spread.¹ Miin takes a different approach: using AI to monitor patterns of account behavior on social platforms in real time, identifying those "coordinated accounts" that simultaneously push messages in the same direction using similar syntax.

During House Speaker Pelosi's visit to Taiwan in 2022, Miin's data showed that one in every four Twitter accounts related to the event was involved in cognitive operations.¹ These accounts did not necessarily publish false news — instead, they manipulated public opinion direction through mass retweeting and amplifying specific narratives. Miin flags these accounts, allowing readers to see traces of manipulation while they read.

📝 Curator's perspective: Taiwan's geographic position makes it a frontline of cognitive warfare. In 2024, Taiwan AI Labs signed memoranda of understanding with two Lithuanian companies specifically focused on AI cognitive warfare solutions.⁵ Miin is not just Taiwan's tool — it is becoming a defensive experience shared internationally.

Tu points out that as deepfake technology matures, "the barrier to creating disinformation with today's AI technology is extremely low."¹ Miin's strategy is not to fact-check item by item — that speed cannot keep pace — but to expose manipulation networks at the level of account behavior.

The Gray Zone of Copyright

But Miin's news aggregation model also stepped on a landmine.

In December 2025, NEXTAPPLE News accused Miin's AI crawler of scraping more than 250 exclusive news articles without authorization, republishing the content on its own platform for users to read and discuss, in suspected violation of copyright law. NEXTAPPLE's parent company Long Cheng Creative has filed a criminal complaint against Ethan Tu.²

📝 Curator's perspective: This is not just Miin's problem — it is a structural contradiction for the entire AI industry. AI needs large amounts of data for training and presentation; news media needs copyright protection to sustain revenue. The copyright lawsuits between Google News, OpenAI, and media companies in various countries are all different battlefields of the same war.

This incident exposed the as-yet-unresolved legal gray zone between "AI aggregation" and "news copyright." Miin claims to be a non-profit open-source platform aimed at improving media literacy⁶; but for news organizations, no matter how noble your purpose, unauthorized use of content is infringement.

As of early 2026, the case remains in judicial process with no final verdict.

The Continuity and Contradiction of the PTT Spirit

From PTT to Miin, Tu's core conviction runs as a through-line: information should be open, transparent, and not controlled by a few. PTT used an anonymous forum to break the media monopoly; Miin tries to use AI to break the algorithmic monopoly.

But over thirty years, the price of "openness" has also escalated. The challenge in the PTT era was how to govern anonymous speech; the challenge in the Miin era is — when you "aggregate" others' content with AI to fight disinformation, are you protecting the public interest, or infringing private property?

There is no simple answer to this question. And Taiwan, as a frontline of disinformation attacks and an experimental ground for AI applications, will probably be one of the first places in the world forced to answer it.

References