Within Community Notes
Can disagreement make notes more trustworthy?
Community Notes depend on agreement across contributors with different rating histories, not just a simple majority vote.
On this page
- What bridging based ranking tries to solve
- Why majority voting can fail in polarized feeds
- How cross viewpoint agreement shapes visibility
Page outline Jump by section
Introduction
Community Notes do not decide visibility by counting votes and declaring a winner. Instead, the system uses a mechanism known as bridging-based ranking, designed to identify notes that people with different rating histories can agree are helpful. The core idea is simple: a note should not become public merely because it is popular within one faction. It should become visible when contributors who often disagree with one another nevertheless converge on the judgment that the note adds useful context. This governance choice is one of the most distinctive features of Community Notes and helps explain why some notes appear beneath posts while many others never become visible. [X (formerly Twitter)]communitynotes.x.comX (formerly Twitter)Note ranking algorithmThe Community Notes ranking algorithm leverages contributor ratings to identify notes that are…
Within the broader challenge of critical thinking on social media, bridging-based ranking is an attempt to reward shared understanding rather than partisan victory. Rather than asking, “Which side has more voters?”, it asks, “Can people who usually disagree still agree that this note helps readers?” [X (formerly Twitter)]communitynotes.x.comX (formerly Twitter)Note ranking algorithmThe Community Notes ranking algorithm leverages contributor ratings to identify notes that are…
What bridging-based ranking tries to solve
Crowdsourced moderation faces a familiar problem: people often evaluate information through political, ideological or community loyalties. If a platform simply publishes the note with the most positive ratings, the largest or most active group can dominate the outcome. In a polarised environment, that may reward factional support rather than informative context. [ACM Digital Library]dl.acm.orgACM Digital LibraryBirds of a feather don't fact-check each other: Partisanship…Here, we leverage data from Birdwatch, Twitter's crowd…
Community Notes was developed partly in response to this problem. Early versions of Twitter’s Birdwatch programme relied more heavily on straightforward helpfulness voting. Researchers and platform designers observed that contributors could rate notes differently depending on whether they aligned with their existing views. Bridging-based ranking was introduced to reduce the influence of those patterns and to elevate notes that attracted approval across divides rather than within a single camp. [arXiv]arxiv.orgFrom Birdwatch to Community Notes, from Twitter to X10 Oct 2025 — To mitigate this issue, Community Notes adopted a bridging algorit…
The goal is not to eliminate disagreement. Instead, it seeks a narrower outcome: identifying notes that diverse contributors independently regard as useful, even when they disagree about many other topics. [X (formerly Twitter)]communitynotes.x.comX (formerly Twitter)Note ranking algorithmThe Community Notes ranking algorithm leverages contributor ratings to identify notes that are…
Why majority voting can fail in polarised feeds
Imagine a note attached to a controversial political post. If 1,000 contributors from one viewpoint rate the note as helpful and 300 contributors from another viewpoint rate it as unhelpful, a simple majority system would likely publish the note.
The difficulty is that such a result does not necessarily show broad confidence in the note’s quality. It may only show that one side outnumbers the other. In highly polarised discussions, people often agree with information that supports their existing beliefs and reject information that challenges them. Research on Birdwatch found measurable partisan patterns in how users evaluated notes and claims. [ACM Digital Library]dl.acm.orgACM Digital LibraryBirds of a feather don't fact-check each other: Partisanship…Here, we leverage data from Birdwatch, Twitter's crowd…
Bridging-based ranking treats this as a signal rather than a nuisance. Instead of counting all ratings equally, it examines historical rating behaviour and looks for evidence that support comes from contributors who do not normally agree with each other. A note that receives approval only from one cluster of contributors may fail to qualify, even if it accumulates many positive ratings. [X (formerly Twitter)]communitynotes.x.comX (formerly Twitter)Note ranking algorithmThe Community Notes ranking algorithm leverages contributor ratings to identify notes that are…
This creates a higher bar than simple popularity. A note must demonstrate cross-group usefulness rather than merely winning a numbers contest. [TechCrunch]techcrunch.comTech Crunch Twitter expands fact-checking program ahead of USIdeally, platforms would then reward behavior that “bridges divides” rather than…Read more…
How cross-viewpoint agreement shapes visibility
The mechanism works by analysing patterns in contributors’ past ratings. Contributors who consistently rate notes in similar ways are grouped closer together, while contributors whose ratings frequently diverge are treated as representing different perspectives. The system does not require contributors to declare political identities; instead, it infers relationships from rating behaviour itself. [X (formerly Twitter)+2arXiv]communitynotes.x.comX (formerly Twitter)Note ranking algorithmThe Community Notes ranking algorithm leverages contributor ratings to identify notes that are…
When a new note receives ratings, the algorithm evaluates more than the total number of “helpful” votes. It asks whether positive ratings come from contributors who historically disagree. A note is more likely to become publicly visible when it attracts support from multiple parts of this rating landscape. Conversely, a note supported only within one cluster may remain unpublished. [X (formerly Twitter)+2The Oversight Board]communitynotes.x.comX (formerly Twitter)Note ranking algorithmThe Community Notes ranking algorithm leverages contributor ratings to identify notes that are…
In practical terms:
- Contributors write a note on a potentially misleading post.
- Other contributors rate whether the note is helpful.
- The system examines both the ratings and the relationships among raters.
- Agreement among historically disagreeing contributors increases the note’s consensus score.
- Notes crossing the required threshold become visible to the public. [X (formerly Twitter)]communitynotes.x.comX (formerly Twitter)Note ranking algorithmThe Community Notes ranking algorithm leverages contributor ratings to identify notes that are…
Importantly, the threshold is not a fixed number of votes. Platform documentation and public explanations indicate that visibility depends on the pattern of agreement, not simply on accumulating a particular count of ratings. [The Oversight Board]oversightboard.compao 007g5zuvThe Oversight BoardAssessing Meta's Plans to Expand Community Notes26 Mar 2026 — Meta explained that the algorithm calculates this score…
Can disagreement make notes more trustworthy?
The theory behind bridging-based ranking is that consensus reached across disagreement carries more informational value than consensus reached within a like-minded group. If people who frequently oppose one another still agree that a note is informative, that agreement may be a stronger signal that the note contains broadly useful context. [ICIP]icip.catBridging Algorithms as Practical Tools for DepolarisationBridging Algorithms as Practical Tools for Depolarisation - ICIPLuke Thorburn explains how “bridge algorithms” work, a digital method…
Research on Birdwatch and Community Notes provides some support for this idea. Studies found that notes selected through bridging-based methods were generally viewed as helpful and reduced users’ willingness to accept or reshare misleading claims. One large study reported that users exposed to algorithm-selected notes were significantly less likely to agree with misleading posts and less likely to reshare them. [arXiv]arxiv.orgarXiv Birdwatch: Crowd Wisdom and Bridging Algorithms canBirdwatch: Crowd Wisdom and Bridging Algorithms can…October 31, 2022 — by S Wojcik · 2022 · Cited by 98 — On average, according t…
The mechanism therefore attempts to turn disagreement into a quality filter. Rather than suppressing conflict, it uses patterns of disagreement to identify instances where common ground emerges despite that conflict. [TechCrunch]techcrunch.comTech Crunch Twitter expands fact-checking program ahead of USIdeally, platforms would then reward behavior that “bridges divides” rather than…Read more…
Limits and ongoing debates
Bridging-based ranking is not a guarantee of truth. A note can fail to appear because contributors from different groups never converge, even if the note is factually accurate. Likewise, some topics are so polarised that cross-group agreement is difficult to achieve. Critics argue that this can create blind spots, particularly for contentious subjects where evidence is strong but consensus is weak. [andyhaupt.com]andyhaupt.comCommunity Notes are Centrist Not ConsensusCommunity Notes are Centrist, Not Consensusby A Haupt · 2025 — We study the relationship between classic models of voting by committees a…
Researchers have also found that consensus can change over time. A 2026 study examining hundreds of thousands of Community Notes reported that a substantial share of displayed notes later lost their helpful status after additional ratings arrived, highlighting the challenge of maintaining stable agreement in contentious environments. [arXiv]arxiv.orgConsensus Stability of Community Notes on XJanuary 20, 2026…
Even so, the mechanism represents a significant departure from conventional moderation systems. Instead of treating agreement as a simple vote count, it treats agreement across disagreement as a scarce and valuable signal. That design choice lies at the heart of how Community Notes decide which corrections readers ultimately see. [X (formerly Twitter)+2Platformer]communitynotes.x.comX (formerly Twitter)Note ranking algorithmThe Community Notes ranking algorithm leverages contributor ratings to identify notes that are…
Amazon book picks
Further Reading
Books and field guides related to Can disagreement make notes more trustworthy?. Use these as the next step if you want deeper reading beyond the article.
The Constitution of Knowledge
Directly addresses how disagreement can produce more reliable knowledge.
Mistakes Were Made (but Not by Me) Third Edition
Explains self-justification and resistance to correction.
eBay marketplace picks
Marketplace Samples
Example marketplace items related to this page. Use the search link to explore similar finds on eBay.
Endnotes
-
Source: platformer.news
Title: meta community notes launch
Link: https://www.platformer.news/meta-community-notes-launch/Source snippet
How Meta's take on Community Notes misses the mark13 Mar 2025 — Community Notes are the most prominent example to date of a product that...
-
Source: icip.cat
Title: Bridging Algorithms as Practical Tools for Depolarisation
Link: https://www.icip.cat/perlapau/en/article/bridging-algorithms-as-practical-tools-for-depolarisation/Source snippet
Bridging Algorithms as Practical Tools for Depolarisation - ICIPLuke Thorburn explains how “bridge algorithms” work, a digital method...
-
Source: dl.acm.org
Link: https://dl.acm.org/doi/10.1145/3491102.3502040Source snippet
ACM Digital LibraryBirds of a feather don't fact-check each other: Partisanship...Here, we leverage data from Birdwatch, Twitter's crowd...
-
Source: arxiv.org
Link: https://arxiv.org/html/2510.09585v1Source snippet
From Birdwatch to Community Notes, from Twitter to X10 Oct 2025 — To mitigate this issue, Community Notes adopted a bridging algorit...
-
Source: techcrunch.com
Title: Tech Crunch Twitter expands fact-checking program ahead of US
Link: https://techcrunch.com/2022/09/07/twitter-expands-its-crowdsourced-fact-checking-program-birdwatch-ahead-of-u-s-midterms/Source snippet
Ideally, platforms would then reward behavior that “bridges divides” rather than...Read more...
-
Source: arxiv.org
Title: arXiv Birdwatch: Crowd Wisdom and Bridging Algorithms can
Link: https://arxiv.org/pdf/2210.15723Source snippet
Birdwatch: Crowd Wisdom and Bridging Algorithms can...October 31, 2022 — by S Wojcik · 2022 · Cited by 98 — On average, according t...
Published: October 31, 2022
-
Source: andyhaupt.com
Title: Community Notes are Centrist Not Consensus
Link: https://www.andyhaupt.com/assets/media/Community_Notes_are_Centrist__Not_Consensus.pdfSource snippet
Community Notes are Centrist, Not Consensusby A Haupt · 2025 — We study the relationship between classic models of voting by committees a...
-
Source: arxiv.org
Link: https://arxiv.org/abs/2601.14002Source snippet
Consensus Stability of Community Notes on XJanuary 20, 2026...
Published: January 20, 2026
-
Source: dl.acm.org
Link: https://dl.acm.org/doi/10.1145/3774904.3792987Source snippet
Stability of Community Notes on X12 Apr 2026 — While prior work has shown that the platform's bridging-based algorithm effectively select...
-
Source: communitynotes.x.com
Link: https://communitynotes.x.com/guide/en/under-the-hood/ranking-notesSource snippet
X (formerly Twitter)Note ranking algorithmThe Community Notes ranking algorithm leverages contributor ratings to identify notes that are...
-
Source: oversightboard.com
Title: pao 007g5zuv
Link: https://www.oversightboard.com/decision/pao-007g5zuv/Source snippet
The Oversight BoardAssessing Meta's Plans to Expand Community Notes26 Mar 2026 — Meta explained that the algorithm calculates this score...
-
Source: Wikipedia
Title: Community Notes
Link: https://en.wikipedia.org/wiki/Community_Notes -
Source: ea-crux-project.vercel.app
Title: community notes
Link: https://ea-crux-project.vercel.app/knowledge-base/responses/community-notes/Source snippet
X Community Notes30 Jan 2026 — LLM Summary:Community Notes uses a bridging algorithm requiring cross-partisan consensus to display fact-c...
Additional References
-
Source: Tech Policy Press
Link: https://techpolicy.press/researchers-explore-partisanship-in-twitter-crowdsourced-fact-checking-program-birdwatchSource snippet
Researchers Explore Partisanship in Twitter Crowdsourced...6 Apr 2022 — Birdwatch allows people to identify information in Tweets they b...
-
Source: semanticscholar.org
Link: https://www.semanticscholar.org/paper/Birdwatch%3A-Crowd-Wisdom-and-Bridging-Algorithms-can-Wojcik-Hilgard/f983a4e27b457a502fba322be1efb675639dde86Source snippet
itative analysis of the efficiency of X's Community Notes, a crowdsourced...Read more...
-
Source: researchgate.net
Title: This approach places Contributors along an opinion
Link: https://www.researchgate.net/publication/396457977_From_Birdwatch_to_Community_Notes_from_Twitter_to_X_four_years_of_community-based_content_moderationSource snippet
ResearchGate(PDF) From Birdwatch to Community Notes, from Twitter to X10 Oct 2025 — To mitigate this issue, Community Notes adopted a bri...
-
Source: hal.science
Title: Bouchaud Ramaciotti 2026 Community Notes v2
Link: https://hal.science/hal-05116614v2/file/Bouchaud%20Ramaciotti%202026%20Community%20Notes%20v2.pdfSource snippet
Community Notes undermoderate polarizing content by...by P Bouchaud · 2025 · Cited by 10 — Our results show that the CNs algorithm effec...
-
Source: jonathanwarden.com
Title: understanding community notes
Link: https://jonathanwarden.com/understanding-community-notesSource snippet
and Bridging-Based Ranking1 Jan 2024 — The Community Notes Documentation explains that the algorithm favors notes that are rated highly b...
-
Source: youtube.com
Title: Mechanics of Community Notes: Bridging Algorithms and Cross-Partisan Consensus
Link: https://www.youtube.com/watch?v=anleNB7b3SUSource snippet
An inside look at X's Community Notes | Keith Coleman & Jay Baxter...
-
Source: youtube.com
Title: Jay Baxter | How did X use bridging-based ranking for community notes?
Link: https://www.youtube.com/watch?v=ow4Fag8P7lASource snippet
Luke Thorburn | What are some of the open questions in bridging systems?...
-
Source: researchgate.net
Link: https://www.researchgate.net/publication/364932548_Birdwatch_Crowd_Wisdom_and_Bridging_Algorithms_can_Inform_Understanding_and_Reduce_the_Spread_of_MisinformationSource snippet
Birdwatch: Crowd Wisdom and Bridging Algorithms can...Bridging algorithms subsequently determine which notes achieve sufficient cross-id...
-
Source: youtube.com
Title: An inside look at X’s Community Notes | Keith Coleman & Jay Baxter
Link: https://www.youtube.com/watch?v=8dgyqYHLcCISource snippet
Recommendations and Democracy: A Way Forward...
-
Source: youtube.com
Title: Recommendations and Democracy: A Way Forward
Link: https://www.youtube.com/watch?v=bR9T1pPFy0MSource snippet
Jay Baxter | How did X use bridging-based ranking for community notes?...
Topic Tree