Authenticity Score Community-Verified
You're only as good as the sources of information you consume. WeBetAI tracks the authenticity of every source on X over time — scored by the community through peer-to-peer challenges on whether their claims turn out to be true. No editorial board. No algorithm deciding what's real. Just people putting their money where their mouth is.
Sources Tracked
2.4K
Claims Challenged
18K+
Resolved Outcomes
12.1K
Avg. Resolution Time
4.2days
The Problem with Information Today
Everyone has an opinion. Nobody has a track record. The accounts you follow, the journalists you read, the influencers you trust — there's no way to know how often they're actually right. A blue check means identity verification, not content authenticity. Follower count measures popularity, not accuracy.
Traditional fact-checking doesn't scale. It's slow, centralized, and inherently subjective. By the time a claim is fact-checked, it's already been shared a million times. And the fact-checker's credibility is just as unverifiable as the original source's.
WeBetAI Authenticity Score solves this with skin in the game. When someone challenges a claim with real money, they're saying "I believe this enough to stake something on it." When thousands of challenges resolve over time, the source's track record writes itself. No editor. No algorithm. Just outcomes.
Authenticity Score Scale
90-100
Trusted
75-89
Reliable
60-74
Mixed
40-59
Caution
0-39
Unreliable
Scores are earned, not assigned. Every point comes from a resolved community challenge. Post a claim. The community challenges it. Time passes. The outcome resolves. Your score adjusts. There is no shortcut, no appeal process, no editorial override. The score is the sum of your track record.
Recency-weighted. Your last 90 days matter more than your first 90. A source that was reliable two years ago but has been posting inauthentic content recently will see their score decline. The score reflects who you are now, not who you were.
Sample Authenticity Leaderboard
Illustrative examples of how scores would look across different source types.
1
@AP
Wire service • 847 claims tracked • 94% confirmed authentic
96
Trusted
2
@BLS_gov
Official data • 312 claims tracked • 98% confirmed authentic
95
Trusted
3
@ShamsCharania
Sports breaking news • 1,204 claims tracked • 91% confirmed
93
Trusted
4
@elikitten
Tech analysis • 189 claims tracked • 82% confirmed
81
Reliable
5
@ViralNewsDaily
Aggregator • 534 claims tracked • 61% confirmed
58
Caution
6
@BreakingAlerts247
Outrage aggregator • 892 claims tracked • 34% confirmed
31
Unreliable
How Authenticity Scoring Works
Challenge Resolution Rate
Primary Signal
The core metric. When a source posts a claim and the community challenges it, the outcome eventually resolves. What percentage of a source's challenged claims turned out to be authentic? This is the backbone of the score. A source with 200 resolved challenges and 180 confirmed authentic has a 90% resolution rate. Simple. Transparent. Unfakeable.
Challenge Volume
Confidence Weight
A source with 5 resolved challenges and 100% accuracy is less reliable than one with 500 challenges and 88% accuracy. Volume gives the score statistical confidence. The more challenges resolved, the more the score converges on the source's true authenticity rate. Bayesian smoothing prevents new sources from starting at 100% with no history.
Recency Weighting
Decay Function
Recent outcomes count more than old ones. A 90-day exponential decay ensures the score reflects current behavior, not historical reputation. A source that was 95% authentic last year but dropped to 60% this quarter will see that reflected immediately. Reputations are earned continuously, not banked.
Claim Severity Weighting
Impact Multiplier
Not all claims are equal. A source that's wrong about a game score loses a few points. A source that's wrong about a health claim, a market-moving event, or a public safety issue loses significantly more. Higher-stakes claims carry higher-stakes consequences for the score. The community determines stakes by how much they're willing to challenge on it.
Source Correction Bonus
Integrity Signal
Sources that correct their own claims before community resolution earn a small score bonus. Self-correction is a signal of integrity. A source that posts something wrong, acknowledges it, and corrects it within 24 hours takes a smaller hit than one that doubles down on inauthentic claims. Accountability is rewarded.
Pipeline Architecture
1
Claim Detection
WeBetAI monitors posts from tracked sources on X. When a post contains a verifiable claim — a specific prediction, a factual assertion, a data point, a "breaking" report — it's flagged as challengeable. Grok's real-time access identifies claims with clear, resolvable outcomes.
AI Detection
2
Community Challenge
Any WeBetAI user can challenge a flagged claim. "I'll put $5 that this turns out to be true" or "I challenge $5 this is wrong." The challenge is peer-to-peer — between two people who disagree on the authenticity of the claim. No house. No spread. Just two people with opposing views and skin in the game.
P2P Challenge
3
Time-Based Resolution
Claims resolve based on their nature: sports claims resolve after the game, economic claims resolve when official data drops, prediction claims resolve at their stated deadline. Resolution is tracked against primary sources — official data, verified outcomes, documented events. Not opinions. Not interpretations. Outcomes.
Outcome-Based
4
Score Adjustment
Each resolved challenge updates the source's Authenticity Score. Confirmed authentic claims strengthen the score. Inauthentic claims weaken it. The magnitude of adjustment depends on challenge volume (confidence), recency (decay), and claim severity (impact). Scores update in real time as outcomes resolve.
Automated
5
Leaderboard + Feed Integration
Source Authenticity Scores are published on the public leaderboard and available via API. Guardian uses these scores to weight sources in the daily digest — higher-scoring sources get more prominence. Over time, your feed naturally gravitates toward authentic voices because the community has done the verification work for you.
Public Record
Why Community Scoring Works Better than Fact-Checking
Skin in the game changes everything. When you put $5 on a claim being authentic, you research it first. You check the source. You look for corroboration. The act of challenging with real stakes turns passive consumers into active investigators. The crowd does the verification work because they're incentivized to.
It scales infinitely. Traditional fact-checking requires paid staff reviewing individual claims. Community challenges happen organically across thousands of claims simultaneously. The more users participate, the more claims get challenged, and the more accurate the scores become. The system gets stronger with every user.
It's immune to editorial bias. No single entity decides what's true. The community decides by putting money on it, and outcomes resolve objectively. A claim is authentic or it isn't. The score is math, not opinion. There's no appeals board, no political pressure, no editorial slant.
It creates accountability where none exists. Right now, a source can post something false, delete it, and move on with zero consequences. With Authenticity Score, every claim is tracked. Every outcome is recorded. The score is a permanent, public record of how often a source tells the truth. That changes behavior.
Mental Health Impact
Reduced Anxiety from Source Uncertainty
High Impact
A huge source of information anxiety is not knowing whether what you're reading is real. When every source has a visible, community-verified Authenticity Score, the uncertainty disappears. You see a 94 next to @AP and a 31 next to @BreakingAlerts247. You know immediately which one to trust. That clarity is a mental health intervention.
Empowered Information Consumption
High Impact
Instead of passively absorbing whatever the algorithm serves, you actively choose your sources based on their track record. You curate your information diet the same way you curate your real diet — by looking at the label. Authenticity Score is the nutrition label for information. That shift from passive consumption to active choice is transformative for how you feel about what you know.
Confidence in Conversations
High Impact
When you share something with a friend and you know the source has a 93 Authenticity Score, you share with confidence. You don't preface it with "I saw this online but I'm not sure if it's real." You say it directly. That confidence transforms conversations and strengthens relationships. You become someone people trust for good information.
Community Connection Through Challenges
Growing
Every challenge is a conversation between two people who care enough about the truth to stake something on it. That's a fundamentally different kind of social interaction than a reply thread or a quote tweet. Challenges build community around shared pursuit of what's authentic — and those connections extend into real life.
Resolution Data Sources
Official Government Data
BLS, Fed, SEC, CDC — primary source for economic and policy claims
Sports Outcomes
ESPN, league APIs — scores, trades, transactions for sports claims
Wire Services
AP, Reuters — primary reporting for news and event claims
Market Data
Stock prices, crypto, prediction markets — for financial claims
WeBetAI Community Consensus
For claims without a definitive primary source, super-majority community resolution (80%+ agreement) with a 48-hour dispute window
Scoring Principles
Outcomes only: Authenticity Score is based on resolved outcomes, not opinions. A claim is authentic or it isn't. The score is math, not editorial judgment.
Transparent methodology: Every component of the score — resolution rate, volume, recency weight, severity multiplier — is documented and auditable. When a source asks "why is my score this number," the answer is a formula, not a committee decision.
No censorship, ever: A low Authenticity Score doesn't hide or suppress a source. It informs the reader. Every source stays visible. Every claim stays challengeable. The score is context, not a content filter.
Self-correction is rewarded: Sources that acknowledge and correct mistakes take a smaller score hit than those who double down or delete. Authenticity isn't about being right every time — it's about being honest about when you're wrong.
The community is the authority: No single entity controls scores. The community challenges, time passes, outcomes resolve, and the score updates. WeBetAI provides the infrastructure. The crowd provides the truth.
Roadmap
P2P challenge infrastructure: READY create-bet.js, send-bet-sms.js, and the full Twilio SMS flow are live in production. The challenge mechanism exists.
X monitoring via Grok: READY search-x.js and grok-chat.js provide real-time claim detection from tracked sources.
Claim extraction pipeline: Automated identification of verifiable claims from source posts. Requires prompt engineering to distinguish claims ("Jobs up 285K") from opinions ("The economy feels strong").
Automated resolution engine: Cross-referencing resolved outcomes against primary data sources (ESPN scores, BLS releases, market prices). Requires API integrations per claim category.
Public leaderboard API: /api/get-authenticity endpoint serving source scores, claim history, and resolution records. Stored in Netlify Blobs with the same pattern as Edge Picks.
Guardian integration: Authenticity Scores feed into Guardian's daily digest weighting. Higher-scoring sources get more prominence. The two systems reinforce each other — Guardian surfaces authentic content, Authenticity Score verifies the sources behind it.