Media Manipulation and Bias Detection
Auto-Improving with AI and User Feedback
HonestyMeter - AI powered bias detection
CLICK ANY SECTION TO GIVE FEEDBACK, IMPROVE THE REPORT, SHAPE A FAIRER WORLD!
Shantilal Mahat (UML candidate) and Deepbahadur Shahi (Congress candidate) are jointly favored as positive examples compared to typical electoral competition.
Caution! Due to inherent human biases, it may seem that reports on articles aligning with our views are crafted by opponents. Conversely, reports about articles that contradict our beliefs might seem to be authored by allies. However, such perceptions are likely to be incorrect. These impressions can be caused by the fact that in both scenarios, articles are subjected to critical evaluation. This report is the product of an AI model that is significantly less biased than human analyses and has been explicitly instructed to strictly maintain 100% neutrality.
Nevertheless, HonestyMeter is in the experimental stage and is continuously improving through user feedback. If the report seems inaccurate, we encourage you to submit feedback , helping us enhance the accuracy and reliability of HonestyMeter and contributing to media transparency.
Presenting a complex or varied reality as simpler or more uniform than it is.
“निर्वाचनमा उम्मेदवारहरूबीच एकआपसमा प्रतिस्पर्धा हुने भएकाले कतिपय स्थानमा बोलचालसमेत हुँदैन । उनीहरूले एकअर्कालाई प्रतिस्पर्धी ठान्ने भएकाले संवाद गर्ने अवस्थासमेत रहँदैन । तर जुम्लामा फरक दृश्य देखिएको छ ।” This suggests that generally candidates do not talk to each other and there is no dialogue, then contrasts Jumla as a ‘different scene’. It simplifies the diversity of electoral behavior across places and times into a near-uniform pattern of non-dialogue.
Qualify the generalization: “धेरै ठाउँमा उम्मेदवारबीच प्रतिस्पर्धाका कारण बोलचाल कम हुने गरेको पाइन्छ...” instead of implying it is almost universal.
Add context or data if available: for example, mention whether this pattern has been observed in previous elections or is based on specific reports, not just implied as a general rule.
Clarify that Jumla is ‘one example’ rather than a unique exception: e.g., “जुम्लामा देखिएको यो दृश्य त्यस्ता सामान्य धारणा भन्दा फरक एउटा उदाहरण हो।”
Drawing a broad conclusion from a small or limited number of examples.
“उनीहरूको यो अभ्यासले जिल्लामा सकारात्मक सन्देश फैलिएको स्थानीयले बताए ।” From a single phone call and a few reactions, the article concludes that a ‘positive message has spread in the district’. This extrapolates district-wide impact from unspecified ‘locals’ without indicating scope or evidence.
Specify the scale more modestly: e.g., “स्थानीय केही बासिन्दाले यसलाई सकारात्मक सन्देशको रूपमा लिएका छन्” instead of “जिल्लामा सकारात्मक सन्देश फैलिएको”.
Attribute the claim more precisely: name or describe the locals (e.g., “स्थानीय नेताहरू”, “बुअ गाउँका केही बासिन्दा”) and indicate that this is their perception, not an established fact.
If available, add supporting evidence (e.g., multiple interviews, public reactions) or remove the broad claim about the entire district.
Using emotionally positive language or framing to encourage a favorable reaction rather than neutrally presenting facts.
“उनीहरूको यो अभ्यासले जिल्लामा सकारात्मक सन्देश फैलिएको स्थानीयले बताए ।” “उनीहरूबीचको यो संवादले निर्वाचनमा हुने हरेक प्रतिस्पर्धा उम्मेदवारबीचको सकारात्मक सोच र आपसी समझदारीसहित अघि बढ्नुपर्ने सन्देश दिएको स्थानीयले बताए ।” These sentences frame the event as carrying a ‘positive message’ and as a normative model for ‘every competition’ to proceed with ‘positive thinking and mutual understanding’. This goes beyond description into value-laden, aspirational framing.
Make clear that these are opinions, not facts: e.g., “स्थानीय केही बासिन्दाको भनाइमा, यस्तो संवादले सकारात्मक सन्देश दिन सक्छ...”
Separate reporting from normative judgment: instead of “हुनुपर्ने सन्देश दिएको”, use “हुनुपर्ने भन्ने धारणा व्यक्त गरेका छन्” to show it is a viewpoint.
Balance with a neutral closing line that simply summarizes what happened, without prescribing how ‘every’ election should be.
Claims presented without sufficient evidence or sourcing detail.
“उनीहरूको यो अभ्यासले जिल्लामा सकारात्मक सन्देश फैलिएको स्थानीयले बताए ।” “...निर्वाचनमा हुने हरेक प्रतिस्पर्धा उम्मेदवारबीचको सकारात्मक सोच र आपसी समझदारीसहित अघि बढ्नुपर्ने सन्देश दिएको स्थानीयले बताए ।” The article cites ‘locals’ but does not specify how many, who they are, or how representative they might be. The scale (‘जिल्लामा’) and normative ‘message’ are asserted without concrete evidence.
Identify the sources more specifically (e.g., “हिमा गाउँपालिकाका केही स्थानीय बासिन्दाले”, “स्थानीय शिक्षक र व्यापारीहरूले”) to give readers context.
Avoid district-wide claims unless supported: change “जिल्लामा सकारात्मक सन्देश फैलिएको” to “स्थानीयले यसलाई सकारात्मक रूपमा लिएका छन्” or similar.
If the claim about a ‘message’ is based on interviews, mention how many people were interviewed or that it is based on limited conversations.
Presenting information in a way that emphasizes a particular interpretation or value judgment.
The structure contrasts a generalized picture of non-communication (“संवाद गर्ने अवस्थासमेत रहँदैन”) with this case as a ‘different scene’ and then closes by saying it gives a message that all electoral competition should proceed with positive thinking and mutual understanding. This framing encourages readers to see this event as exemplary and normatively desirable, rather than simply one instance among many possible campaign behaviors.
Reframe the contrast more neutrally: e.g., “सामान्यतया कडा प्रतिस्पर्धा हुने निर्वाचनमा यस्तो संवाद कम देखिन्छ, तर जुम्लामा यस्तो एउटा उदाहरण देखिएको छ।”
In the conclusion, describe rather than prescribe: “यस घटनालाई केही स्थानीयले उम्मेदवारबीचको आपसी समझदारीको उदाहरणका रूपमा व्याख्या गरेका छन्” instead of stating what ‘should’ happen in every competition.
Optionally, include that not all observers may interpret it the same way, or simply end with factual description of the phone call and reactions.
- This is an EXPERIMENTAL DEMO version that is not intended to be used for any other purpose than to showcase the technology's potential. We are in the process of developing more sophisticated algorithms to significantly enhance the reliability and consistency of evaluations. Nevertheless, even in its current state, HonestyMeter frequently offers valuable insights that are challenging for humans to detect.