Tuesday, 27 October 2026 Updated hourly from 226 outlets
Pricing Sign in

4 sourcesFirst reported yesterday at 08:16, updated 4 h ago

In China, the AI bubble has already burst

Photo: bdnews24.com

Neutral summary

from 3 outlets' reporting

China's mainland STAR50 index of AI and semiconductor leaders has fallen 40 percent in less than four months, wiping out about $480 billion in market value, according to Reuters-sourced reports. The slump has halted a two-year rally that began with economic stimulus promises and stalled Beijing's push for a "slow bull" market. Traders cited high valuations, tax bills and the threat of US or other countries banning Chinese equipment in data centres, though there was no clear trigger. Goldman Sachs data show AI stocks make up 42 percent of China's market capitalisation. Fund manager Dong Baozhen said "now we see the bubble bursting," while Hong Hao of Lotus Asset Management said the selloff would halt many listings. The China Securities Regulatory Commission did not immediately respond to questions. Some investors see value in the selloff, while others, such as fund manager Wang Zhuo, expect a price war.

Where accounts differ. The Kathmandu Post names the broad blue-chip index as the CSI 300, while bdnews24.com and The Financial Express refer only to China's broad blue-chip index, so the naming is not fully consistent across outlets.

The Kathmandu PostThe Financial Expressbdnews24.com

Written by AI from these outlets' articles: facts they report, claims attributed, numbers checked against the sources. It may still contain errors; read the originals.

Article analysis

3out of 10

Band BLocalised faults

The Kathmandu PostStated as fact

Headline adopts the 'bubble burst' framing voiced by quoted fund managers; narration is largely data-driven with mild evaluative wording and includes contrary investor views and a regulator non-response.

Zero of the four outlets that covered this event headlined it neutrally.

The faults are local; the reporter's narration stays descriptive.

Read the originalSee the working

1 Straight reporting10 Campaigning

The same event, one framing

The earliest headline of each kind. Marked words are judgements in the headline itself: stated as fact or in quotation marks.

Stated as fact 4 of 4

In China, the AI bubble has already burst

The Express Tribune

How the 3 was worked out

The model answers seven fixed questions from 0 to 3 and cites evidence for each. The score is calculated from the answers in code, so any disagreement points to one question.

  1. G1
    Reporter's own voiceOne isolated evaluative word in the reporter's voice. Evidence 2 3
    1
  2. G2
    HeadlineCompresses or intensifies, but the claim stays attributed. Evidence 1
    1
  3. G3
    Other side missingEvery party involved responds substantively.
    0
  4. G4
    Weight of the unanswered allegationNo allegation against an identifiable party.
    0
  5. G5
    Loaded quotesSome evaluative words in quotes, in proportion to the event. Evidence 4
    1
  6. G6
    Unattributed claimsOne or two disputable claims are unattributed. Evidence 5
    1
  7. G7
    Missing contextNecessary background is present.
    0

Evidence

1Headline-body attribution mismatchHeadline voice

In China, the AI bubble has already burst

Headline asserts in the paper's voice the 'bubble burst' characterization that the body attributes to fund managers, though the 40% fall supports a crash framing.

2Word choiceReporter voice

lofty valuations collapse

Evaluative 'lofty' and dramatic 'collapse' in narration, though backed by cited multiples.

3Word choiceReporter voice

The market frenzy

'Frenzy' is an evaluative characterization in the reporter's voice.

4Word choiceAttributed quote voice

speculative crazy bull

Charged language from a quoted fund manager, proportionate market commentary.

5Unsubstantiated claimReporter voice

has pushed many investors towards defensive assets

Claim about investor behaviour stated without a source.

Show the calculation
S = G2 + … + G7 = 1 + 0 + 0 + 1 + 1 + 0 = 3
G1 = 1, below 2: not band D
Not band C or A: band B
Band B: S ≤ 3 gives 3, 4 to 6 gives 4, 7 or more gives 5. Score 3

Every headline, in order

  1. The Express Tribune South Asia First to report Stated as fact

    In China, the AI bubble has already burst

  2. The Kathmandu Post South Asia Analysed above Stated as fact

    In China, the AI bubble has already burst

  3. bdnews24.com Bangladesh Stated as fact

    In China, the AI bubble has already burst

  4. The Financial Express Bangladesh Stated as fact

    In China, the AI bubble has already burst

About this method

Method. Bias types follow the Table of Media Bias Elements and the Media Bias Taxonomy (both CC BY 4.0). The impartiality rules follow the BBC Editorial Guidelines and the Ofcom Broadcasting Code. This analysis was made by claude-opus-5-5 on rubric 1.0.0. The score is calculated from the seven answers, never generated.

What this is not. No outlet gets a standing bias rating, and there is no left/right axis, because that axis does not describe Bangladeshi politics. Each analysis covers one article about one event. Members' analytics count how each outlet's headlines were labelled over a chosen period, with the number of headlines behind every share, and also show the average score of an outlet's scored articles: that average describes those articles (mostly each story's lead), not the outlet, and outlets are never ranked by it or by any share.

Known limits. Headlines are grouped into stories automatically, so an unrelated headline sometimes slips in; it is shown faded rather than hidden. Where outlets carry the same wire copy, the comparison has little to show. Human annotators agree on bias in Bangla news only moderately (κ = 0.574 at best), which limits what any automated score can claim.

The full method