kahawatungu.com Media Bias



Empirical bias analysis of the supplied article sample (41 records, Oct 2026)

Topic selection: The sample splits into four clusters:   Kenyan police-blotter crime/court reporting — roughly a third of items cover stabbings, body discoveries, mob justice, and remand hearings                           ;   formulaic celebrity net-worth profiles               ;   Kyen political/business coverage — Ruto opposition attacks   , Kindiki reelection claims   , Kenya Pipeline earnings   , Kenya Airways–Safaricom promotion   , Ebola response     ;   international/world news that reads as BBC-style wire copy — Falklands oil dispute   , Pope Leo health   , Anthropic results   , Netflix layoffs   , US inflation   , Starlink/Ukraine   , livestreamed execution     .

AI authorship (probabilistic, not certain): Multiple records contain strong textual artifacts of automated drafting from a supplied dataset: repeated 'supplied information/supplied source' leakage in net-worth pieces         , plus 'prompt-leak phrasing indicating automated drafting'   and 'template-generation artifacts' with contradictory figures ($16M vs $6M)   .

The verbatim BBC passages         suggest syndicated third-party copy alongside templated generated content.

Framing patterns: Crime coverage is consistently attribution-heavy, hedged, and police-reliant ('according to police,' 'reportedly,' pending postmortems)         , with rare lapses like a headline asserting a father's guilt beyond hedged body text   and DCI 'breakthrough' framing adopted uncritically   .

The most evident ideological lean is establishment/pro-incumbent on Kenyan politics: Ruto's attacks on the opposition run unchallenged with 'demonstrating the Government's commitment' narration   , Kindiki's development claims appear without opposition response   , and the Ebola piece frames government response as decisive with appeal-to-authority sourcing   , while an op-ed piece issues 'should' directives favoring state-led health authority   . A counterexample of critical framing exists: the execution articles lean critical of the Trump administration via expert/anonymous sources     , and Burnham's transparency gap is spotlighted   .

Reliability pattern: Strong procedural attribution and hedging (integrity bias noted in       ) but low independent verification — single official sources dominate crime/health items     , corporate promotion is relayed uncritically     , and net-worth figures are consistently unsourced speculation       .

Highest 3 values: procedural accuracy/attribution     ; speed/frequency of local crime coverage     ; institutional officialdom as primary truth-source     .
Lowest 3 values: editorial independence/critical scrutiny       ; sourcing rigor of financial claims     ; fact-checking consistency (conflicting figures, leaked prompts)     .

Propaganda techniques observed: appeal to authority     , promotional/advertising relay     , establishment framing     (UK-tilted Falklands framing).

No counterexample to the pro-incumbent Kenya tilt appears in the sample.


Helium Bias: Analysis rests on one 41-item sample biased toward recent, unusual/high-bias articles; I cannot observe unpublished stories.

AI-authorship is inferred from textual artifacts, not confirmed.

Source name unknown; BBC-style passages may be syndication rather than authorship.

Topic-count categorization involves judgment calls.

Automated source summary · Updated October 18, 2026 · Not human reviewed. Check recent article panels for claim-level evidence when available.




Use the Data in AI All Sources

kahawatungu.com Bias Profile

Weighted source-level patterns from recent analyzed coverage. Open recent articles below to inspect score-specific evidence and limitations when available.

🗞️ Objective <—> Subjective 👁️ -21

💡 Boring <—> Interesting12

😨 Fearful16

📞 Begging the Question6

🗣️ Gossip16

💭 Opinion50

🗳 Political34

Oversimplification18

🏛️ Appeal to Authority40

👀 Covering Responses11

😢 Victimization16

😤 Overconfidence14

🗑️ Spam17

🔒 Ideological24

🏴 Anti-establishment <—> Pro-establishment 📺12

❌ Low Credibility <—> High Credibility ✅22

🧠 Rational <—> Irrational 🤪-19

🤑 Advertising21

🤖 Written by AI78

💔 Low Integrity <—> High Integrity ❤️22

🪨 Low Intelligence <—> High Intelligence 🦉28

🎭 Virtue Signaling18

🔍 Truth-seeking <—> Delusion 🌀-20

🔬 Scientific <—> Superstitious 🔮-15

🎲 Speculation17

🐍 Manipulative35

Subtle dimensions

🔵 Liberal <—> Conservative 🔴-1

🧢 Populist <—> Elitist 🎩2

🗽 Libertarian <—> Authoritarian 🚔0

🚨 Sensational0

📉 Bearish <—> Bullish 📈5

😩 Pessimistic <—> Optimistic 🌞4

📝 Prescriptive0

🕊️ Dovish <—> Hawkish 🦁4

🍼 Immature4

🔄 Circular Reasoning2

📏📏 Double Standard4

💣 Terrorism2

✊ Woke5

🔺 Conspiracy0

🐐 Scapegoating2

🤡 Hypocrisy2

🦊 Anti-Corporate <—> Pro-Corporate 👔3

👤 Individualist <—> Collectivist 👥1

How to interpret source scores →

Average social shares per article 0





kahawatungu.com Recent Articles



Sort By: