India's Cockroach Janta Party protests were not an organic student movement but a foreign-backed influence operation, run with Pakistani bot followers, US funding, and an opposition toolkit

Why the suspicion had a foothold

The people who found this movement suspicious were not all acting in bad faith, and the case for looking twice is worth stating properly before it is answered.

Start with the curve. A satirical website became a mass political formation with a larger Instagram following than the ruling party in roughly three weeks. Anyone who has tried to organise anything knows that is not how organising usually goes. When growth outruns every intuition about how movements build, “someone is running this” is not a paranoid thought; it is the first hypothesis a reasonable person reaches for.

Then the people. The founder was a professional political communications strategist, not a student. The chief spokesperson was a journalist. Reporting and social-media digging established that several of the movement's public faces had previous political associations. None of that is disqualifying, and none of it is secret, but a movement presenting itself as spontaneous youth anger while being fronted by experienced political operators does invite the question of who is steering.

A satirical site out-followed the governing party in three weeks. That is a genuinely strange fact, and asking how is the right instinct. The answers offered are where this goes wrong.

And India has real precedent for organised online campaigns. Coordinated networks, purchased engagement, and astroturfed hashtags are documented features of the country's information environment, deployed across the political spectrum. A public trained by that environment to assume manufacture is not being irrational. It is generalising from experience.

So the question was fair. What follows is not a defence of the movement; it is what happened when the specific answers people gave to that fair question were checked.

The numbers that did not hold

The load-bearing claim was arithmetical, which is why it broke first. On 23 May, Minister of State for Education Sukanta Majumdar told reporters, in remarks carried by Press Trust of India, that 49 percent of the CJP page's followers were from Pakistan and only 9 percent from India. He did not say where the other 42 percent were. The same day, Union Minister of Parliamentary Affairs Kiren Rijijuposted that he pitied those who sought their followers from “Pakistan & George Soros gang,” and Kerala BJP president Rajeev Chandrasekhar described the movement as a cross-border influence operation designed to destabilise India.

Three measurements say otherwise, and they come from three different directions. Dipke published the account's own Instagram audience breakdown, showing a little over 94 percent of followers located in India. One of the movement's X accounts posted a screen recording of its reach data putting the figure at 94.7 percent. And The Quint ran its own independent analysis using the audience-analytics platform Modash, which again found the large majority in India and only a small share identifiable as bots.

Platform analytics are estimates, and a fair critic can say so. But the asymmetry is stark. On one side, three readings taken by different parties using different tools, all landing within a couple of points of each other. On the other, a percentage stated at a press conference with no methodology, no source, and 42 percent of the audience unaccounted for. Those are not competing datasets. One is data and the other is an assertion.

The secret-controller claim collapsed just as cleanly. In June, users found that an X account posting about the movement had older posts referring to two Samajwadi Party politicians as parents, and concluded the opposition had built the CJP. The date stamps in the very screenshots being shared read 2022 and 2023, before the CJP existed and before the Chief Justice made the remark that prompted it. The Quintchecked the account's history: created in 2022, username changed 53 times. That is the signature of a recycled account jumping onto a trending cause, a thing that happens constantly, and not of a founder caught out.

When the proof is synthetic

The second thing that happens when you check this narrative is that its corroborating media keeps turning out to be manufactured. This matters more than any individual fake, because a claim can be true while its worst advocates lie. What it cannot easily survive is having every piece of confirming evidence dissolve on contact.

A deepfake of External Affairs Minister S Jaishankar circulated with the false claim that he had asked the United States to deport Dipke. A video was altered to make the YouTuber Dhruv Rathee appear to have exposed the movement for buying followers, an accusation he never made. Another was edited to suggest Shah Rukh Khan was joining. A clip of the Chief of Defence Staff, General NS Raja Subramani, spread with audio in which he seemed to say that certain cockroaches in India would be treated like pests; the original livestream by the news agency ANI contains nothing of the kind. An AI-generated letter was produced to suggest a foreign political organisation was funding the protest. An AI-manipulated photograph placed the activist Anna Hazareat a press conference he did not attend. Old footage from unrelated workers' protests was recirculated as CJP crowds.

The toolkit was described but never shown. The letter proving foreign backing was generated. The minister demanding a deportation was a deepfake. At some point the pattern is the finding.

The toolkit claim deserves its own line, because it is the purest case. A video circulating before the 6 June protest asserted that viral leaks and screenshots revealed a step-by-step blueprint for manufacturing national outrage. The blueprint was never produced. The claim travelled as a description of evidence rather than as evidence, which is a durable format precisely because there is nothing to examine.

One item resists this treatment and should be kept out of the pile. The Ministry of Electronics and Information Technology did ask X to withhold a CJP account inside India, citing, per The Indian Express, Intelligence Bureau inputs describing it as a threat to the sovereignty of India and noting its traction among the young. That is a real action on a stated basis, and Dipke went to the Delhi High Court over it. But a restriction is not a finding. No supporting material has been published, the movement's spokesperson said the account carried only satire, protest images, and calls for peaceful protest, and reading an unexplained administrative act as confirmation of the foreign-operation story mistakes an assertion for a verdict.

Why this story was always going to be told

Strip out the specifics and the foreign-hand narrative has a shape that existed long before this movement did. It is worth naming, because the shape is what made it deployable within eight days of a satirical website going up.

The furniture was already in the room. Foreign funding, an opposition toolkit, and a shadowy financier are standing fixtures of Indian political argument online, reached for across the spectrum whenever a movement becomes inconvenient. The George Soros formulation in particular is not an Indian invention: it is a globally recycled trope with a documented antisemitic lineage, in which a single Jewish financier is cast as the hidden purchaser of otherwise inexplicable popular anger. Slotting a new movement into that frame requires no research at all, which is exactly why it is fast.

The second driver is authority. Had the foreign-hand claim stayed with anonymous accounts it would have been background noise. It was voiced by a Union minister, a minister of state, and a state party president, with a specific percentage attached. Most people do not audit a minister citing a statistic; they update. The correction arrived later, in a fact-check article, to a much smaller room.

The third is the asymmetry of effort. “49 percent from Pakistan” is one line. Establishing that the real figure is near 94 percent required pulling account analytics, obtaining a screen recording, and commissioning an independent audience analysis. That work was done, and it still cannot travel at the speed of the sentence it answers.

And underneath all of it sits a genuinely uncomfortable position for a government: a movement it could not easily characterise, making a demand that was eventually conceded. If the protest was foreign, the concession was strategy. If the protest was real, the concession was a defeat. The narrative was doing work, and work is the best predictor of what gets repeated.