Skip to main content
AVOID.NET

Audit log

Every state-changing event for MEV Bot Scam — YouTube AI Trading Bot Campaign (2026): moderation decisions on community submissions, plus corrections and updates from the news pipeline. URL-based decisions are designed to carry three independent witnesses — the original source, an Internet Archive snapshot, and a Solana memo signed by our publicly-disclosed publisher key. Archive coverage is still being backfilled, so each decision below reports its own snapshot status rather than assuming one exists.

  1. #1publishby system:backfill
    2026-08-14 23:26:15Z
    Score: ? → ? (no score change)
    anchoranchored
    chain
    ●mainnet-betaslot 443,507,704
    sig
    4xqjULZ6YaNu…yTJnws8Hexplorer ↗
    hash
    2jujYrUwQ7Df…PTYi7Rijsha256 → base58
    verifying row…full verify ↗
    canonical bytes (21805 B) ▸
    {"actor":"system:backfill","investigation_id":"7b08e7ad-67e1-442e-b82a-62c08f9922d7","kind":"publish","page_slug":"mev-bot-scam-youtube-ai-trading-bot-campaign-2026","published_at":"2026-08-14T23:26:15.260Z","sequence_num":1,"snapshot":{"content_type":"investigation","entity_name":"MEV Bot Scam — YouTube AI Trading Bot Campaign (2026)","sections":[{"content":"SentinelLABS, the threat research division of SentinelOne, published a detailed report on August 5, 2025 documenting an active campaign in which malicious smart contracts are promoted as Maximal Extractable Value (MEV) arbitrage bots through YouTube tutorial videos. The campaign exploits the legitimate appeal of MEV bot strategies — real MEV bots can generate substantial returns by front-running trades — to lure retail crypto users seeking passive income. Victims are instructed to deploy the provided Solidity code via the Remix IDE, fund the resulting smart contract with a minimum of 0.5 ETH, then call a function such as 'Start()' or 'StartNative()' to supposedly activate the bot. Instead, funds are immediately redirected to attacker-controlled wallets via obfuscated backdoors embedded in the contract code. The campaign primarily targets Ethereum mainnet users. Academic research published by San Diego State University covering the period June 2022 through June 2023 identified 10,442 arbitrage bot scam videos across YouTube and estimated that the broader scam family caused $7.13 million to $15 million in losses to approximately 25,000 victims across Ethereum and Binance Smart Chain.","heading":"Campaign Overview","severity":"critical","sources":[{"credibility":1,"name":"SentinelOne Labs: Smart Contract Scams — Ethereum Drainers Pose as Trading Bots to Steal Crypto","type":"research","url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"},{"credibility":2,"name":"CoinDesk: Weaponized Trading Bots Drain $1M From Crypto Users via AI-Generated YouTube Scam","type":"news_article","url":"https://www.coindesk.com/tech/2025/08/07/weaponized-trading-bots-drain-usd1m-from-crypto-users-via-ai-generated-youtube-scam"},{"credibility":1,"name":"SDSU Academic Study: Towards Understanding and Characterizing the Arbitrage Bot Scam In the Wild (arXiv)","type":"research","url":"https://arxiv.org/html/2310.12306v1"}]},{"content":"SentinelLABS identified three distinct obfuscation techniques used by the malicious contracts to conceal the attacker's destination wallet address from casual code inspection. First, an XOR method XORs two 32-byte constants assigned misleading names such as 'DexRouter' and 'factory' to derive the attacker wallet address at runtime. Second, a string concatenation method assembles the attacker address from fragments that appear innocuous in isolation. Third, a large decimal-to-hex conversion casts 256-bit integers to the 'uint160' data type to reconstruct wallet addresses without presenting them in recognizable hexadecimal form. Additionally, the contracts include a failover mechanism: even if a victim never calls the 'Start()' function, the attacker retains the ability to withdraw any deposited ETH unilaterally, ensuring the scam extracts funds regardless of victim behavior. SentinelLABS identified the following SHA-1 hashes for contract samples tied to the documented campaign: 2923cdf2caba3a92e0ea215d14343ce73e8f08a5 (Jazz_Braze variant), f0a34770f03428c8abc9e73df93263f10f8320b1 (SolidityTutorials variant), and 47d567e799f0403bcd4057bff50244125cac926a (ToddTutorials variant). Malicious code was hosted on short-link URLs including pastebin.com/raw/8Yar7QyU, is.gd/SvLp4d, tinyurl.com/m89fj9wm, and codeshare.io/0bV94e.","heading":"Technical Mechanism: Obfuscated Backdoors","severity":"critical","sources":[{"credibility":1,"name":"SentinelOne Labs: Smart Contract Scams — Ethereum Drainers Pose as Trading Bots to Steal Crypto","type":"research","url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"}]},{"content":"A defining characteristic of this campaign is the use of aged YouTube accounts to bypass platform trust signals. SentinelLABS noted that the @Jazz_Braze account, for example, had published approximately 100 pop culture videos and shorts between November 2022 and April 2024 before pivoting to the scam tutorial, which accumulated more than 387,000 views. Scammers acquire aged accounts through Telegram marketplaces at prices ranging from approximately $6 to thousands of dollars. Once an account is deployed for the scam, operators actively manage the YouTube comment section by deleting negative feedback and posting fabricated positive reviews to sustain an appearance of legitimacy. Many videos are distributed as unlisted content via Telegram or direct messages rather than through public search, reducing exposure to platform moderation. Multiple videos in the campaign exhibited indicators of AI-generated content: robotic voiceover cadence, inconsistent speech pacing, twitchy or choppy facial expressions, and lip movements misaligned with audio. SentinelLABS noted that the Jazz_Braze video was an exception and appeared to use a human narrator, suggesting the operators vary their production approach to evade detection. Blockchain security firm SlowMist separately documented, in October 2024, an evolution of the same scam in which operators rebranded from 'Uniswap Arbitrage MEV Bot' to 'ChatGPT Arbitrage MEV Bot,' appending AI branding to exploit public interest in large language models and reduce victim skepticism.","heading":"Distribution Infrastructure: Aged YouTube Accounts and AI-Generated Content","severity":"high","sources":[{"credibility":1,"name":"SentinelOne Labs: Smart Contract Scams — Ethereum Drainers Pose as Trading Bots to Steal Crypto","type":"research","url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"},{"credibility":2,"name":"CoinTelegraph: Old MEV bot scam rides AI hype to return with new name — SlowMist","type":"news_article","url":"https://cointelegraph.com/news/old-mev-bot-scam-drops-chatgpt-into-new-name-ai-hype-slowmist"},{"credibility":2,"name":"crypto.news: Malicious crypto trading bots siphon over $900k via aged YouTube accounts","type":"news_article","url":"https://crypto.news/malicious-crypto-trading-bots-siphon-over-900k-via-aged-youtube-accounts/"}]},{"content":"SentinelLABS documented three specific YouTube channels and their associated attacker externally-owned account (EOA) addresses as of the August 2025 report. The @Jazz_Braze channel was associated with wallet 0x872528989c4D20349D0dB3Ca06751d83DC86D831, which collected approximately 244.9 ETH (roughly $902,000 USD). The @SolidityTutorials channel was linked to 0x7359EA6AA3343b3238171e76F97e6aA3cDB8d696, which collected approximately 4.19 ETH (roughly $15,000 USD). The @todd_tutorials channel (video subsequently made private) was associated with 0x2eEE3A0ed51AE22813050212E8D9Fa216bd51df4, which collected approximately 7.59 ETH (roughly $28,000 USD). Funds from the Jazz_Braze primary EOA were subsequently distributed to at least 24 additional recipient addresses, complicating on-chain tracing. SentinelLABS estimated total documented losses across identified contracts exceeded $1 million. The New Zealand Financial Markets Authority (FMA) separately published a public warning about suspected scam videos promoting cryptocurrency sniping bots and MEV bots on YouTube. The San Diego State University academic study, covering June 2022 through June 2023 across a broader set of 1,697 scam addresses and 10,442 videos, estimated losses of $7.13 million to $15 million and approximately 25,000 victims across Ethereum and Binance Smart Chain for that earlier period, with victims losing an average of $490 to $1,000 each on Ethereum.","heading":"Financial Losses and On-Chain Evidence","severity":"critical","sources":[{"credibility":1,"name":"SentinelOne Labs: Smart Contract Scams — Ethereum Drainers Pose as Trading Bots to Steal Crypto","type":"research","url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"},{"credibility":2,"name":"CoinDesk: Weaponized Trading Bots Drain $1M From Crypto Users via AI-Generated YouTube Scam","type":"news_article","url":"https://www.coindesk.com/tech/2025/08/07/weaponized-trading-bots-drain-usd1m-from-crypto-users-via-ai-generated-youtube-scam"},{"credibility":1,"name":"SDSU Academic Study: Towards Understanding and Characterizing the Arbitrage Bot Scam In the Wild (ACM/arXiv)","type":"research","url":"https://arxiv.org/html/2310.12306v1"},{"credibility":1,"name":"New Zealand FMA: Suspected scam — videos promoting cryptocurrency sniping and MEV bots","type":"regulatory","url":"https://www.fma.govt.nz/library/warnings-and-alerts/suspected-scam-videos-promoting-cryptocurrency/"},{"credibility":2,"name":"web3isgoinggreat.com: Traders lose $1 million to malicious trading bot software","type":"news_article","url":"https://www.web3isgoinggreat.com/?id=mev-bot-scam"}]},{"content":"SlowMist published findings in October 2024 documenting a deliberate rebranding strategy. The scam, originally marketed as 'Uniswap Arbitrage MEV Bot,' was repackaged as 'ChatGPT Arbitrage MEV Bot' to exploit public interest in AI tools. SlowMist observed that by associating the fraudulent product with OpenAI's ChatGPT brand, operators were able to increase perceived credibility and reduce victim skepticism. SlowMist's analysis identified at least two scammer addresses that had collectively stolen 30 ETH (approximately $78,000) from over 100 victims and 20 ETH (approximately $52,000) from 93 victims respectively since August 2024. The operators' distribution strategy intentionally targets low-to-mid sized deposits to minimize any single victim's incentive to pursue recovery, spreading losses across a large number of targets rather than concentrating exposure. This 'wide net' approach makes collective attribution and law enforcement action more difficult. As of the date of this investigation (August 2026), no criminal charges, indictments, or regulatory enforcement actions specifically targeting the operators of this campaign have been identified in public records. The New Zealand FMA issued a consumer warning, but no formal enforcement action was documented. No charges or convictions have been entered against any identified operator. All characterizations of the actors as fraudulent are based on on-chain evidence and security researcher analysis.","heading":"Campaign Evolution and AI Rebranding","severity":"high","sources":[{"credibility":2,"name":"CoinTelegraph: Old MEV bot scam rides AI hype to return with new name — SlowMist","type":"news_article","url":"https://cointelegraph.com/news/old-mev-bot-scam-drops-chatgpt-into-new-name-ai-hype-slowmist"},{"credibility":2,"name":"SlowMist Medium: A New Twist on Old Scams — The Arbitrage MEV Bot Deception","type":"research","url":"https://slowmist.medium.com/a-new-twist-on-old-scams-the-arbitrage-mev-bot-deception-8a9118b4645a"},{"credibility":2,"name":"DailyCoin: SlowMist Exposes New Scamming Trend in Arbitrage MEV Bots","type":"news_article","url":"https://dailycoin.com/slowmist-flags-new-breed-of-mev-scam-bots-targeting-crypto-users/"}]},{"content":"The context submitted with this investigation claims the campaign targets both Ethereum and Solana users. The primary SentinelLABS August 2025 report documents only Ethereum mainnet targeting through Solidity smart contracts deployed via Remix IDE. No Solana-specific variant of this particular campaign (involving Solidity backdoor contracts promoted via AI YouTube videos) was corroborated by SentinelLABS or SlowMist reporting as of the research conducted for this page. A separate, distinct campaign documented by JFrog Security Research involved malicious npm and PyPI packages ('Solana FakeFix') that asked developers to paste a Solana private key — this is a different attack vector and distribution method from the YouTube MEV bot campaign. The academic SDSU study additionally identified Binance Smart Chain as a secondary target network alongside Ethereum. Any claim that the specific YouTube AI trading bot campaign described by SentinelLABS has expanded to Solana should be treated as low-confidence until corroborated by additional security research.","heading":"Solana Targeting Claims","severity":"medium","sources":[{"credibility":1,"name":"SentinelOne Labs: Smart Contract Scams — Ethereum Drainers Pose as Trading Bots to Steal Crypto","type":"research","url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"},{"credibility":2,"name":"JFrog Security Research: Solana FakeFix — Malicious npm and PyPI Packages","type":"research","url":"https://research.jfrog.com/post/solana-fakefix/"}]},{"content":"SentinelLABS recommends that users avoid deploying smart contract code obtained from YouTube tutorials or other social media sources, regardless of the view count or apparent legitimacy of the hosting account. Users should independently verify any code through reputable security auditors before funding a deployed contract. Deployment of any contract on Ethereum mainnet should be preceded by review on a testnet using only worthless test ETH. Comment sections on YouTube are not reliable indicators of legitimacy, as operators actively suppress negative feedback and fabricate positive reviews. Aged YouTube accounts with histories in unrelated content categories (pop culture, news) are not in themselves indicators of trustworthiness. Any tutorial requiring a minimum ETH deposit to 'activate' a trading bot should be treated with strong skepticism. Legitimate MEV bots are operated by sophisticated teams with access to significant capital and infrastructure; they are not distributed as public tutorials.","heading":"Defensive Guidance","severity":"low","sources":[{"credibility":1,"name":"SentinelOne Labs: Smart Contract Scams — Ethereum Drainers Pose as Trading Bots to Steal Crypto","type":"research","url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"},{"credibility":2,"name":"CCN: Fake Trading Bots on Ethereum Fuel $900K Smart Contract Scam","type":"news_article","url":"https://www.ccn.com/news/technology/ethereum-fake-trading-bots-smart-contract-scam/"}]}],"sources_used":[{"credibility":1,"name":"SentinelOne Labs: Smart Contract Scams — Ethereum Drainers Pose as Trading Bots to Steal Crypto","type":"research","url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"},{"credibility":2,"name":"CoinDesk: Weaponized Trading Bots Drain $1M From Crypto Users via AI-Generated YouTube Scam","type":"news_article","url":"https://www.coindesk.com/tech/2025/08/07/weaponized-trading-bots-drain-usd1m-from-crypto-users-via-ai-generated-youtube-scam"},{"credibility":1,"name":"SDSU Academic Study: Towards Understanding and Characterizing the Arbitrage Bot Scam In the Wild (arXiv)","type":"research","url":"https://arxiv.org/html/2310.12306v1"},{"credibility":2,"name":"CoinTelegraph: Old MEV bot scam rides AI hype to return with new name — SlowMist","type":"news_article","url":"https://cointelegraph.com/news/old-mev-bot-scam-drops-chatgpt-into-new-name-ai-hype-slowmist"},{"credibility":2,"name":"SlowMist Medium: A New Twist on Old Scams — The Arbitrage MEV Bot Deception","type":"research","url":"https://slowmist.medium.com/a-new-twist-on-old-scams-the-arbitrage-mev-bot-deception-8a9118b4645a"},{"credibility":2,"name":"crypto.news: Malicious crypto trading bots siphon over $900k via aged YouTube accounts","type":"news_article","url":"https://crypto.news/malicious-crypto-trading-bots-siphon-over-900k-via-aged-youtube-accounts/"},{"credibility":1,"name":"New Zealand FMA: Suspected scam — videos promoting cryptocurrency sniping and MEV bots","type":"regulatory","url":"https://www.fma.govt.nz/library/warnings-and-alerts/suspected-scam-videos-promoting-cryptocurrency/"},{"credibility":2,"name":"web3isgoinggreat.com: Traders lose $1 million to malicious trading bot software","type":"news_article","url":"https://www.web3isgoinggreat.com/?id=mev-bot-scam"},{"credibility":2,"name":"CCN: Fake Trading Bots on Ethereum Fuel $900K Smart Contract Scam","type":"news_article","url":"https://www.ccn.com/news/technology/ethereum-fake-trading-bots-smart-contract-scam/"},{"credibility":2,"name":"Security Boulevard: Scams Using Malicious Smart Contracts, YouTube videos to Steal Crypto","type":"news_article","url":"https://securityboulevard.com/2025/08/scams-using-malicious-smart-contracts-youtube-videos-to-steal-crypto/"},{"credibility":2,"name":"JFrog Security Research: Solana FakeFix — 25 Malicious npm and PyPI Packages","type":"research","url":"https://research.jfrog.com/post/solana-fakefix/"},{"credibility":2,"name":"DailyCoin: SlowMist Exposes New Scamming Trend in Arbitrage MEV Bots","type":"news_article","url":"https://dailycoin.com/slowmist-flags-new-breed-of-mev-scam-bots-targeting-crypto-users/"},{"credibility":1,"name":"ACM: Towards Understanding and Characterizing the Arbitrage Bot Scam In the Wild (peer-reviewed)","type":"research","url":"https://dl.acm.org/doi/10.1145/3626783"}],"summary":"An ongoing campaign, active since at least mid-2022 and continuing through 2025, uses AI-generated YouTube videos to promote malicious Solidity smart contracts disguised as Maximal Extractable Value (MEV) arbitrage trading bots. According to SentinelOne Labs (SentinelLABS), one attacker wallet alone collected approximately 244.9 ETH (roughly $902,000 USD) from victims, with total documented losses across the broader campaign exceeding $1 million. Victims are deceived into deploying backdoored contracts via Remix IDE and depositing ETH, which is then routed directly to attacker-controlled wallets through obfuscated code.","timeline":[{"date":"2022-06-01","event":"Earliest documented arbitrage bot scam videos identified on YouTube, per SDSU academic study covering June 2022 through June 2023.","source":"SDSU Academic Study (arXiv)","source_url":"https://arxiv.org/html/2310.12306v1"},{"date":"2022-11-01","event":"YouTube account @Jazz_Braze begins uploading approximately 100 pop culture videos and shorts, apparently to establish an aged account with apparent credibility before pivoting to the scam.","source":"SentinelOne Labs","source_url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"},{"date":"2023-06-01","event":"End of the SDSU study period. By this date the academic researchers had identified 10,442 scam videos, 1,697 scam addresses, and estimated losses of $7.13 million to $15 million across approximately 25,000 victims.","source":"SDSU Academic Study (arXiv)","source_url":"https://arxiv.org/html/2310.12306v1"},{"date":"2024-01-01","event":"Nefture Security research reported that in the first three months of 2024 alone, approximately $2 million was lost by nearly 1,000 victims, indicating the campaign continued to accelerate after the SDSU study period.","source":"Nefture Security via web3isgoinggreat.com","source_url":"https://www.web3isgoinggreat.com/?id=mev-bot-scam"},{"date":"2024-06-06","event":"@Jazz_Braze publishes the video 'How to Create Passive Income MEV Bot on Ethereum Full Tutorial,' which goes on to accumulate over 387,000 views and is linked to attacker wallet 0x872528989c4D20349D0dB3Ca06751d83DC86D831, which collected approximately 244.9 ETH (~$902,000 USD).","source":"SentinelOne Labs","source_url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"},{"date":"2024-08-01","event":"SlowMist documents new MEV bot scam activity beginning around this date, including operators pivoting from 'Uniswap Arbitrage MEV Bot' branding to 'ChatGPT Arbitrage MEV Bot' to exploit AI market enthusiasm.","source":"SlowMist via CoinTelegraph","source_url":"https://cointelegraph.com/news/old-mev-bot-scam-drops-chatgpt-into-new-name-ai-hype-slowmist"},{"date":"2024-10-13","event":"SlowMist publishes its findings on the rebranded 'ChatGPT Arbitrage MEV Bot' scam, identifying at least two attacker addresses that had collectively stolen at least 50 ETH (~$130,000 USD) from over 190 victims since August 2024.","source":"CoinTelegraph / SlowMist","source_url":"https://cointelegraph.com/news/old-mev-bot-scam-drops-chatgpt-into-new-name-ai-hype-slowmist"},{"date":"2025-04-01","event":"SentinelLABS identifies the most recent scam contract it analyzed in this campaign, indicating the operation was still active as of April 2025.","source":"SentinelOne Labs","source_url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"},{"date":"2025-08-05","event":"SentinelOne Labs publishes its full report documenting the campaign, naming three YouTube channels (@Jazz_Braze, @SolidityTutorials, @todd_tutorials), three attacker wallets, and three contract SHA-1 hashes. Total documented losses reported as exceeding $1 million.","source":"SentinelOne Labs","source_url":"https://www.sentinelone.com/labs/smart-contract-scams-ethereum-drainers-pose-as-trading-bots-to-steal-crypto/"},{"date":"2025-08-07","event":"CoinDesk publishes coverage of the SentinelLABS report, reporting total losses as exceeding $1 million and describing the AI-generated video production methods.","source":"CoinDesk","source_url":"https://www.coindesk.com/tech/2025/08/07/weaponized-trading-bots-drain-usd1m-from-crypto-users-via-ai-generated-youtube-scam"}]},"v":1}
    Verify offline (run on your own machine)
    python -m src.verify_decision 379618c1-5ee3-49bd-aaae-fed2701fb689
  2. #2reviewby reviewerreviewer
    2026-08-25 13:44:18Z
    Score: 2 → 2 (no score change)
    The page's technical and channel/wallet-specific claims are unusually well-supported and precisely match the primary SentinelOne/SentinelLABS report (exact hash, address, and URL matches), and it appropriately pushes back on an unsupported Solana-targeting claim rather than adopting it. The main problems are a recurring overstatement of total losses (page attributes 'exceeding $1 million' to SentinelLABS in three places, though SentinelLABS's own figure is 'more than $900k'), a minor address-count error in SlowMist-sourced figures, and one timeline entry (Q1 2024 'Nefture Security' loss figure) whose cited source does not contain the claim and whose firm attribution does not match any source found.
    anchoranchored
    chain
    ●mainnet-betaslot 443,514,712
    sig
    2Zga9j6tNmpz…1TTs1jUWexplorer ↗
    hash
    7tdt7mUBsKMD…KSGxWGAnsha256 → base58
    verifying row…full verify ↗
    canonical bytes (1090 B) ▸
    {"actor":"reviewer","decided_at":"2026-08-25T13:44:17.758Z","decision":"review","investigation_id":"7b08e7ad-67e1-442e-b82a-62c08f9922d7","new_score":2,"page_slug":"mev-bot-scam-youtube-ai-trading-bot-campaign-2026","prev_score":2,"reason":"The page's technical and channel/wallet-specific claims are unusually well-supported and precisely match the primary SentinelOne/SentinelLABS report (exact hash, address, and URL matches), and it appropriately pushes back on an unsupported Solana-targeting claim rather than adopting it. The main problems are a recurring overstatement of total losses (page attributes 'exceeding $1 million' to SentinelLABS in three places, though SentinelLABS's own figure is 'more than $900k'), a minor address-count error in SlowMist-sourced figures, and one timeline entry (Q1 2024 'Nefture Security' loss figure) whose cited source does not contain the claim and whose firm attribution does not match any source found.","score_delta":0,"sequence_num":2,"submission_content_hash":null,"submission_id":null,"submission_kind":null,"submission_valence":null,"v":1}
    Verify offline (run on your own machine)
    python -m src.verify_decision 821f3ad6-a74a-48c7-a08b-cc5c15192c34
  3. #3review reviseby judgejudge
    2026-08-25 13:44:18Z
    Score: 2 → 0 (-8)
    By raw count this page checks out well: 28 of 34 claims are confirmed and only 1 is disputed (2.9%), and the technical detail — wallet addresses, contract hashes, obfuscation code, and short-link URLs (claim_findings[9], [10], [11]) — matches the primary SentinelOne report exactly. The page also correctly declined to adopt an unsupported claim that the scam targets Solana and correctly kept a separate JFrog-documented campaign distinct from this one (claim_findings[24], [25]), which are signs of careful editing. But the one disputed item is not a minor slip: a timeline entry attributes a '$2 million, ~1,000 victims' Q1 2024 figure to 'Nefture Security' via a cited URL that contains no such claim, no mention of that firm, and no matching figures at all (claim_findings[28]) — this looks like an invented source attribution rather than a wrong date or number. Separately, the page states in three places that SentinelLABS itself put total losses at 'exceeding $1 million,' when SentinelOne's own report says 'more than $900k' and the $1M figure traces only to a news outlet's rounded headline (claim_findings[3], [18], [32]) — a repeated misattribution of a number to the wrong source. Neither problem touches the page's core, well-supported allegations, but together they are enough unresolved sourcing integrity issues to warrant correction before full approval.
    anchoranchored
    chain
    ●mainnet-betaslot 443,514,733
    sig
    miC8yyHvyKMW…q58wrLT9explorer ↗
    hash
    Cui2m4TxLUw8…hdNQZS83sha256 → base58
    verifying row…full verify ↗
    canonical bytes (1760 B) ▸
    {"actor":"judge","decided_at":"2026-08-25T13:44:17.758Z","decision":"review_revise","investigation_id":"7b08e7ad-67e1-442e-b82a-62c08f9922d7","new_score":0,"page_slug":"mev-bot-scam-youtube-ai-trading-bot-campaign-2026","prev_score":2,"reason":"By raw count this page checks out well: 28 of 34 claims are confirmed and only 1 is disputed (2.9%), and the technical detail — wallet addresses, contract hashes, obfuscation code, and short-link URLs (claim_findings[9], [10], [11]) — matches the primary SentinelOne report exactly. The page also correctly declined to adopt an unsupported claim that the scam targets Solana and correctly kept a separate JFrog-documented campaign distinct from this one (claim_findings[24], [25]), which are signs of careful editing. But the one disputed item is not a minor slip: a timeline entry attributes a '$2 million, ~1,000 victims' Q1 2024 figure to 'Nefture Security' via a cited URL that contains no such claim, no mention of that firm, and no matching figures at all (claim_findings[28]) — this looks like an invented source attribution rather than a wrong date or number. Separately, the page states in three places that SentinelLABS itself put total losses at 'exceeding $1 million,' when SentinelOne's own report says 'more than $900k' and the $1M figure traces only to a news outlet's rounded headline (claim_findings[3], [18], [32]) — a repeated misattribution of a number to the wrong source. Neither problem touches the page's core, well-supported allegations, but together they are enough unresolved sourcing integrity issues to warrant correction before full approval.","score_delta":-8,"sequence_num":3,"submission_content_hash":null,"submission_id":null,"submission_kind":null,"submission_valence":null,"v":1}
    Verify offline (run on your own machine)
    python -m src.verify_decision 7fac2443-6a6b-49eb-bc5c-42bb4a1fe53e
  4. #4reviewby reviewerreviewer
    2026-10-17 07:27:00Z
    Score: 0 → 0 (no score change)
    The page is substantially well-sourced: wallet addresses, ETH amounts, SHA-1 hashes, obfuscation techniques, short-link URLs, the SDSU academic figures, and the JFrog/Solana caveat were all independently verified byte-for-byte or near-exactly against primary sources (SentinelOne Labs, arXiv/SDSU, JFrog). Three defects were found: (1) a recurring overstatement, in the summary, a section, and a timeline entry, attributing an 'exceeding $1 million' total loss figure directly to SentinelLABS, when SentinelOne's own report states 'more than $900k' (the $1M figure is CoinDesk's framing, correctly attributed only in one other timeline entry); (2) a miscount of SlowMist's scammer addresses as 'two' in two places when SlowMist's own report and CoinTelegraph's coverage describe three; and (3) a timeline entry citing Nefture Security for a $2 million/Q1-2024 figure that does not appear in the cited source (web3isgoinggreat.com) and could not be corroborated anywhere else despite Nefture being a real, searchable firm -- this entry warrants the most scrutiny as a possible fabrication or misattribution.
    anchoranchored
    chain
    ●mainnet-betaslot 457,953,437
    sig
    5piQgRi2HNLk…ZtdBSjUwexplorer ↗
    hash
    6D32bzDidaJR…GbH1562Rsha256 → base58
    verifying row…full verify ↗
    canonical bytes (1739 B) ▸
    {"actor":"reviewer","artifact_identity":"6835b5e6d85a3234c8dd8dc0b0c714dc","decided_at":"2026-10-17T07:26:59.858Z","decision":"review","disputed_pct":0.1667,"findings_hash":"48go5xqyKHqQ1yiUNVhQ3XZwqBqRBPyzYRQCg9uVhZcR","investigation_id":"7b08e7ad-67e1-442e-b82a-62c08f9922d7","mode":"findings_only","new_score":0,"page_content_hash":"7f1149a88c27bb2a7c84eb05baba63d7b23965731d6b6de6833050f4fea4ddff","page_slug":"mev-bot-scam-youtube-ai-trading-bot-campaign-2026","prev_score":0,"reason":"The page is substantially well-sourced: wallet addresses, ETH amounts, SHA-1 hashes, obfuscation techniques, short-link URLs, the SDSU academic figures, and the JFrog/Solana caveat were all independently verified byte-for-byte or near-exactly against primary sources (SentinelOne Labs, arXiv/SDSU, JFrog). Three defects were found: (1) a recurring overstatement, in the summary, a section, and a timeline entry, attributing an 'exceeding $1 million' total loss figure directly to SentinelLABS, when SentinelOne's own report states 'more than $900k' (the $1M figure is CoinDesk's framing, correctly attributed only in one other timeline entry); (2) a miscount of SlowMist's scammer addresses as 'two' in two places when SlowMist's own report and CoinTelegraph's coverage describe three; and (3) a timeline entry citing Nefture Security for a $2 million/Q1-2024 figure that does not appear in the cited source (web3isgoinggreat.com) and could not be corroborated anywhere else despite Nefture being a real, searchable firm -- this entry warrants the most scrutiny as a possible fabrication or misattribution.","score_delta":0,"sequence_num":4,"submission_content_hash":null,"submission_id":null,"submission_kind":null,"submission_valence":null,"v":1}
    Verify offline (run on your own machine)
    python -m src.verify_decision 5943d9b9-8f1f-433a-b401-41ba8b421923
How verification works. The “Row integrity” check above is computed in your browser — your machine recomputes the SHA-256 of the canonical bytes and compares against the stored hash. No avoid.net server can fake that check. The “full verify” link goes one level deeper: your browser fetches the on-chain transaction from a Solana RPC node and confirms the same hash is in the memo. If you don’t want to trust either avoid.net or the public RPC, run the CLI verifier on your own machine — python -m src.verify_decision <event_id>.