Download
← All fact checksAfternoon run · October 31, 2026 · #6 by reach
ConflictPoliticsAI-generated media

Did Lt Gen Girish Kalia say China has taken control of Asaphila, Longju and Taksing in Arunachal Pradesh?

Did Lt Gen Girish Kalia say China has taken control of Asaphila, Longju and Taksing in Arunachal Pradesh, as a viral video claims?

False

Real 2023 footage, fake AI voice: Kalia never said China holds these Arunachal areas.

False. The video is a deepfake. The Indian Army's fact-check handle says AI lip-syncing and voice cloning were laid over genuine footage from 27 October 2023, when Girish Kalia — then a Major General in Kupwara — spoke about counter-infiltration operations, not China. BOOM, Jansatta and Loksatta reached the same conclusion.

+5Sources ↓Original post ↗
Deepfake
The post as it circulated, labelled by our media check.Video · 0:33 · 1280×720

Claims

✕ 3 false

China has taken control of Asaphila, Longju and Taksing circle in Arunachal Pradesh.136

✕ FalseCentral claimHigh confidence

This is the claim the whole post rests on, and it fails at the source: the officer in the video did not say it. The Indian Army's fact-check handle examined the clip and stated that the remarks were fabricated using AI lip-syncing and voice cloning over real archival footage, and that Kalia made no reference to China, Arunachal Pradesh or supply shortages. Fact-checkers BOOM, Jansatta and Loksatta independently traced the frames back to an ANI post dated 27 October 2023 — a briefing in Kupwara, Jammu and Kashmir, about counter-infiltration operations along the Line of Control. That gives us two independent lines of evidence: an official source with access to the officer, and newsrooms that matched the visuals to dated original footage. The second line matters because it does not depend on the Army's word. Beyond the video, no reporting anywhere describes Chinese control of these three circles. Reporting around the same period mentioned sporadic face-offs at some Arunachal locations, which is a very different thing from territorial control. If a serving corps commander had said this publicly, it would have led national news; the only coverage is debunking.

The Army has been requesting supplies from the central government but is instead being told to deploy paramilitary forces against Gen Z in Arunachal Pradesh rather than against China.124

✕ FalseHigh confidence

This line is part of the same synthetic script. The tell is structural rather than semantic: it is stitched into footage that, according to the Army and three newsrooms, was recorded in October 2023 in Kupwara, in a briefing about counter-terror operations in Jammu and Kashmir. Nothing in the genuine briefing concerns supplies, China or protesters. Republic World and Open Magazine both describe the original as a press briefing on counter-infiltration operations along the Line of Control, with the officer then a Major General commanding the 28 Infantry Division. That detail matters because the post's framing depends on him speaking as a current corps commander about current events. A reader should also weigh the sourcing: the Army and the fact-checkers are not wholly independent of each other on the archival-match point, but the archival match is independently checkable by anyone with the 2023 ANI video, and BOOM's AI voice detector scored the audio 18 out of 100 for authenticity — a separate, technical line of evidence pointing the same way.

Twenty-five soldiers were arrested yesterday and their whereabouts are unknown.67

✕ FalseHigh confidence

An arrest of 25 soldiers would be a major, near-impossible-to-suppress story, and it appears in no reporting we found. Instead it sits only inside the AI-generated script laid over the 2023 footage, and the Army confirmed the officer made no such statement. This claim is worth studying because it is the kind of detail that makes a fake feel credible: specific, emotional and hard to check quickly. The specificity is the hook — a round number plus a vague "we don't know their whereabouts" invites outrage rather than verification. Jansatta's and Asianet Newsable's accounts of the Army statement both make clear that the arrest line forms part of the fabricated remarks, not a separate real event wrapped in a fake frame. Our searches for the underlying event returned nothing but coverage of the deepfake itself. When a claim describes an event that thousands of people would witness and dozens of newsrooms would report, silence across the whole press is itself evidence.

What was posted

X

A post on X shares a short video presented as a statement by Lieutenant General Girish Kalia, General Officer Commanding of the Indian Army's Spear Corps. In the clip the officer appears to say China has taken control of three areas in Arunachal Pradesh, that troops are being denied supplies and pointed at young protesters instead, and that 25 soldiers were arrested the day before. The caption calls the situation "Shameful!" and asks where the accountability is.

“Shameful! Lt Gen Girish Kalia (GOC, Spear Corps) says China has taken control of Asaphila, Longju and Taksing Circle in Arunachal Pradesh. Despite requests for supplies, forces are reportedly being directed against Gen Z instead of China. Where is the accountability?”

Posted on X by an account using the name "India Genz". The clip carries a "Baba Thoka" watermark also linked to other AI deepfakes of Indian Army officers, and the same footage was shared by Congress leader Ritu Choudhary with similar wording, according to The Times of India and BOOM.

Who posted it

X
@indiagenzzz
View the account on X ↗

Account check

Looks like a genuine account4%
Likelihood of automation90%

Estimates from public signals — account age, posting pattern, audience — not a judgement of the person behind the account, and separate from whether this post is true.

What stood out

  • Handle @IndiaGenzzz was named by the Indian Army's official fact-check account (Oct 28, 2026) among 10 handles pushing manipulated videos targeting the Army Chief and soldiers in Arunachal Pradesh.
  • Part of a copy-paste amplification cluster (with @MarkhorTweet, @NilakshiRabhaIN, @Clashfire, @NetaNamaX, @sumitsarkr and others) pushing identical text about the Lachung fire incident, per NewsBharati (Oct 24, 2026).
  • Described as using 'curated Gen-Z aesthetics, sleek profile pictures' with a name designed to pass off as a mainstream Indian user — a hallmark of sock-puppet networks.
  • Cited by Alt News and The Quint (Oct 25–26, 2026) for sharing an edited ANI video falsely claiming Lt Gen Girish Kalia admitted China controls parts of Arunachal; the post got ~850k views.
  • Reportedly part of a Pakistan-linked puppet network previously flagged for trying to infiltrate domestic political protests in New Delhi.
  • Profile is anonymous (no display name, bio, follower count, or creation date disclosed) and not platform-verified; sample of recent posts empty, so no organic sourcing behaviour can be observed.
  • Rapid, near-simultaneous posting in a coordinated window (00:40–08:00 on Oct 28, 2026) alongside nine other handles indicates inauthentic coordination rather than individual commentary.
Signals (9)+
Posting cadence
Burst coordination within narrow overnight windows alongside 9 other handles
Platform verified
No
Anonymous profile
Yes
Notable news coverage
Indian Army fact-check (Oct 28, 2026); Alt News; The Quint WebQoof; Organiser; Bharat Express; NewsBharati
Fact check history
Multiple false/edited-media claims: fabricated Kalia quote on Arunachal (Alt News, The Quint); misleading Lachung fire narrative (NewsBharati); manipulated COAS video (Indian Army)
Coordination evidence
Named in two separate lists of a coordinated disinformation cluster sharing identical copy-paste text
State actor link
Alleged Pakistan-based puppet network per Indian Army fact-check and media reports
Reach
Single post reported at ~850,000 views
Recent posts sample size
0

Account checked October 31, 2026, 14:58 UTC

The media

Video
DeepfakePresented as real footage· The claim depends on it

✓ What is genuine

The footage itself is real: an ANI-recorded press briefing from 27 October 2023, in which the officer spoke about foiled infiltration attempts, terrorists eliminated and seizures of weapons and narcotics along the Line of Control. The man, his uniform and his name tag are genuine — only the words are not.

The tells

  • ●●● Strong signal The visuals match genuine ANI footage from 27 October 2023 in Kupwara; only the words are new.the Indian Army and BOOM
  • ●●● Strong signal Lip movements and audio were synthesised over the original, producing a fabricated statement.the Indian Army fact-check
  • ●●● Strong signal An AI voice detector scored the audio 18/100 for authenticity.BOOM Fact Check
  • ●●○ Supporting signal A "Baba Thoka" watermark links the clip to an account previously associated with AI deepfakes of Army officers.BOOM Fact Check
  • ●●○ Supporting signal The officer is described in the post by his 2026 rank while the footage is from 2023, when he was a Major General.our visual review

Synthetic-media detector

Detector ran

The claim stands or falls on whether the media is real, so the detector ran.

AI-generated likelihood0%
Deepfake likelihood49%

Scores are the detector’s estimate for the file as uploaded. A low score does not rule out editing, a spliced caption, or footage shown out of context — those are judged above. Model v3.

How we checked the media

  • Reviewed the clip's visuals, watermark and on-screen name tag against the archive footage described by the Army and fact-checkers.
  • Ran an automated deepfake check on the video; it returned a deepfake score of 0.49, which on its own is inconclusive.
  • Compared our reading with the Indian Army fact-check and with BOOM, Jansatta and Loksatta, all of which matched the frames to dated 2023 footage.

Sources

8 sources reviewed

Further reading

  • 1BOOM Fact Check BOOM's fact-check sets out the archival match and the AI voice-detection score in detail.
  • 6Jansatta Jansatta walks through the reverse image search that pinned the footage to October 2023.
  • 7Asianet Newsable (ANI) The Army's own fact-check post gives the official explanation and the original 2023 statement.
  • 5The Times of India The Times of India covers the political reaction, showing how the clip travelled across the spectrum.

The full story

The clip is built on something real, which is why it looks convincing. On 27 October 2023, ANI recorded a press briefing in Kupwara, Jammu and Kashmir, by Girish Kalia, then a Major General commanding the 28 Infantry Division. In that briefing, according to the Army's fact-check and the fact-checkers who traced the footage, he discussed counter-infiltration operations along the Line of Control: ten infiltration attempts foiled, 27 terrorists eliminated, and seizures of weapons and narcotics.

None of that appears in the viral version. Instead, an AI system re-voiced the officer and matched his lips to a new script about China seizing Asaphila, Longju and Taksing, about supplies being denied, and about 25 soldiers arrested. The Indian Army's fact-check handle published its response on 25 October 2026, describing the technique as synthetic generated information — AI lip-syncing and voice cloning overlaid on genuine archive video.

There is a thin thread of real context underneath the panic. The ITBP chief has acknowledged recent face-offs with Chinese troops at some locations in Arunachal Pradesh. Face-offs at a border are not the same as one country taking control of named circles, and no source supports the control claim. That gap between a real, tense border situation and a fabricated admission is where the post does its work — the background makes the fake line feel like something that could plausibly be said.

Kernel of truth

Border tensions with China in Arunachal Pradesh are real and have been acknowledged, including by the ITBP chief; the post inflates sporadic face-offs into territorial control and puts the words in an officer's mouth.

Timeline

  1. October 27, 2023

    ANI records then Major General Girish Kalia's press briefing in Kupwara, Jammu and Kashmir, on counter-infiltration operations along the Line of Control. 2Republic World

  2. October 25, 2026

    The Indian Army's fact-check handle says the viral clip is AI lip-syncing and voice cloning overlaid on that 2023 footage, and that Kalia never mentioned China, Arunachal Pradesh or supply shortages. 7Asianet Newsable (ANI)

  3. October 26, 2026

    The clip circulates widely, including a share by Congress leader Ritu Choudhary, and fact-checkers publish their findings. 5The Times of India

How it was framed

Manipulated contentFabricated content

Genuine footage of a real officer was altered with AI lip-syncing and voice cloning to make him say things he never said. The words themselves were invented, so the post is both manipulated media and fabricated content, but the manipulation of real footage is the defining feature.

  • Emotional language. "Shameful!" opens the caption, and the video asks "What message is our Centre government giving to our armed jawans?"
  • Borrowed authority. The claim is placed in the mouth of a named serving officer, "Lt Gen Girish Kalia (GOC, Spear Corps)".
  • “Visual proof”. A video of a uniformed officer with a "GIRISH KALIA" name tag is offered as the evidence itself.
  • Ongoing-threat framing. "China has taken control of Asaphila, Longju and Taksing Circle" and "25 soldiers have been arrested yesterday"
  • Us vs. them. The script contrasts soldiers facing China with forces "directed against Gen Z".

Posted on X by an account using the name "India Genz". The clip carries a "Baba Thoka" watermark that BOOM links to an X handle previously associated with AI deepfakes of Indian Army officers. The same footage was shared by Congress leader Ritu Choudhary, and The Times of India reports that Union minister Kiren Rijiju demanded an apology from the Congress over the shares.

Bias

NoneSlightModerateStrong

Leans partisan opposition attack on the central government. The post is framed as an accountability attack ('Where is the accountability?') built entirely on a fabricated quote, pairing an invented admission of territorial loss with an invented claim that forces are being diverted against young protesters. It pushes a strongly partisan, anti-government narrative with no factual basis.

Bias describes how the post frames its message — the slant of the language, selection and emphasis — and is judged separately from whether the claim is true. A true claim can still be framed with a strong slant.

How we checked

37 s
  1. 1

    Searched for the claim and the officer's name, and for the Indian Army's fact-check response.

  2. 2

    Reviewed the clip's visuals and watermark against the archive footage described by the Army and fact-checkers.

  3. 3

    Ran an automated deepfake check on the video.

  4. 4

    Compared our findings with the Indian Army fact-check and with BOOM, Jansatta and Loksatta.

✓ Claim checked✓ Media checked✓ Sources checked

Why 95% confidence

High

HIGH because two independent lines of evidence agree: an official statement from the Army, which has direct access to the officer, and newsroom verification that matched the frames to dated 2023 footage. The fact-checkers also used an AI voice detector that scored the audio 18/100. We are not relying on a single source or on the detector alone — which is just as well, since our own detector returned a deepfake score of 0.49, squarely inconclusive.

Limits. We did not open the underlying pages in full; our comparison relies on source snippets and the accounts they give of the original footage. We could not independently confirm who created the manipulated clip or when it was first posted.

What we don’t know

  • Who first created the manipulated clip, and whether the "Baba Thoka" watermark belongs to the creator or a reposter.
  • When exactly the manipulated version first appeared, before the 25-26 October 2026 wave of shares.
  • What the ITBP chief's acknowledgement of face-offs in Arunachal covered in detail, beyond the fact that it did not amount to control of these areas.

Spot it yourself

Spot the fake

Looks suspicious, isn’t proof

  • The account name "India Genz" and the partisan framing are reasons to slow down, but plenty of angry posts are accurate.
  • The "Shameful!" tone and the demand for accountability are rhetorical, not evidence either way.
  • A watermark like "Baba Thoka" looks shady, but watermarks are added by aggregators too and prove nothing on their own.
  • The clip looks polished and broadcast-quality — "too perfect" is not a test either; the problem here is the opposite, that the pictures are real.

The actual evidence

  • The Indian Army's fact-check handle stated that the remarks were AI lip-synced and voice-cloned over real footage and never made.
  • Fact-checkers matched frames to an ANI post dated 27 October 2023 — the footage predates the claim by nearly three years.
  • BOOM's AI voice detector scored the audio 18/100 for authenticity.
  • No reporting anywhere supports Chinese control of Asaphila, Longju or Taksing; the only coverage is debunking.
  • No reporting of 25 soldiers being arrested — an event that could not stay quiet.
  1. 1

    Ask when the footage was really recorded

    Search the officer's name and the locations named, then look for fact-checks. Here a single search surfaced the October 2023 date, which already breaks the post's timeline.

  2. 2

    Compare the clip to the original

    The Army's fact-check and BOOM both link to or describe the original 2023 briefing. Watch it: same room, same uniform, same officer, different words.

  3. 3

    Inspect the mouth and the audio

    AI lip-syncing often produces slightly blurred edges around the lips and a voice that sounds flat or oddly paced. Freeze the frame on a close-up and listen with headphones.

  4. 4

    Check whether the event would be huge news

    A serving corps commander saying China has seized Indian territory would lead every bulletin. If only social posts carry it, that silence is evidence.

  5. 5

    Trace the watermark

    Search the watermark's name. Here it led fact-checkers to an account already linked to deepfakes of Army officers — useful context, though never proof by itself.

Takeaway

A video can show a real person and still be fake; the pictures can be genuine while the words are invented. Before believing what someone "says" on camera, establish when the footage was recorded and find the original.

Questions people ask

Is the Girish Kalia video real?+

No. The footage is real but the words are not. The Indian Army's fact-check handle says AI lip-syncing and voice cloning were used to lay a fabricated script over genuine video from 27 October 2023, when Kalia briefed reporters in Kupwara about counter-infiltration operations in Jammu and Kashmir.

What did Girish Kalia actually say in the original 2023 video?+

According to the Army and fact-checkers, the 2023 briefing covered counter-infiltration operations along the Line of Control, including foiled infiltration attempts, terrorists eliminated and seizures of weapons and narcotics. Nothing in it concerned China, Arunachal Pradesh, supplies or arrested soldiers.

Has China taken control of Asaphila, Longju or Taksing?+

No source supports that. Reporting around the same period described sporadic face-offs between Indian and Chinese troops at some Arunachal locations, which the ITBP chief has acknowledged. Face-offs at a border are not the same as one country taking control of named areas.

How can I tell a deepfake of a public figure from a real video?+

Find the original. Search the person's name with the location or phrase they supposedly used, watch for a dated original, and compare. Look at whether the event would have been widely reported, inspect close-ups of the mouth, and check any watermark. One clue alone rarely settles it; agreement across several does.

Were 25 Indian soldiers arrested, as the video says?+

There is no reporting of any such arrest. The line is part of the AI-generated script laid over the 2023 footage, and the Army confirmed the officer made no such statement. An arrest of 25 soldiers would be heavily reported, and the only coverage of this claim is debunking.

Go deeper

Deepfake
Video, audio or images in which AI has altered a real person's appearance or voice so they appear to say or do something they never did.
Voice cloning
Technology that learns the sound of a person's voice from recordings and generates new speech in that voice.
Lip-syncing (AI)
Software that redraws a person's mouth movements in existing video to match a new audio track.
General Officer Commanding (GOC)
The senior officer in charge of a division or corps in the Indian Army.
Line of Control
The de facto border between Indian- and Pakistani-administered Kashmir.

People also search for

Girish Kalia China Asaphila Longju Taksing videoLt Gen Girish Kalia Spear Corps deepfakeShameful Lt Gen Girish Kalia China Arunachal PradeshKalia 25 soldiers arrested China claimIndian Army MythbusterXX Kalia video fact check
For educators+
Deepfakes and synthetic media: real footage can be altered without the pictures looking fake.Borrowed authority: claims gain weight when placed in the mouth of an official.Absence of evidence: for major events, the lack of reporting is itself informative.Provenance: establishing when and where media was first published.
  • The video shows a real officer in a real room. Why does that make the fake harder to spot than a fully generated video would be?
  • If a serving corps commander had said China had taken three areas of Arunachal Pradesh, what would you expect to see in the news within a day?
  • Who benefits from putting these words in an officer's mouth, and what does that suggest about how such clips spread across political lines?

About this check

Platform
X
Posted by
@indiagenzzz
Posted
October 25, 2026
Age when checked
Older post, still live
Where it spread
India, China
Audience
International audience
Language
EN
Status
● Post is live
Verified
October 31, 2026, 15:00 UTC
Published
October 31, 2026, 15:00 UTC

Reach. 11.7K likes, 609 replies; Indian Army debunk; part of ongoing Arunachal deepfake wave

This verdict was generated with AI from the sources shown. AI can make mistakes — check the sources before relying on it.

Check anything yourself, in seconds.

Share any social post or video to What the Fact and get the verdict, the confidence, and the receipts.

Download on theApp StoreGet it onGoogle Play