{"paper_id":"2311.18717","title":"Can AI Detect Wash Trading? Evidence from NFTs","authors":["Brett Hemenway Falk","Gerry Tsoukalas","Niuniu Zhang"],"paper_date":"2023-11-30","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2311.18717v3","page_url":"https://thequant.space/flowcharts/can-ai-detect-wash-trading-evidence-from-nfts/","scores":{"empirical_rigor":8,"math_complexity":6.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes an AI‑based estimator that combines roundedness regressions with machine‑learning to detect NFT wash trading and reduce estimation errors.","quote":"we develop an AI-based estimator that integrates these regressions in a machine learning framework, significantly reducing both exchange- and trade-level estimation errors in NFT markets (and beyond).","summary":"Existing studies on crypto wash trading often use indirect statistical methods or leaked private data, both with inherent limitations. This paper leverages public on-chain NFT data for a more direct and granular estimation. Analyzing three major exchanges, we find that ~38% (30-40%) of trades and…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Filters 1‑3 are extremely conservative and unlikely to misclassify organic behavior as wash trading.","type":"model","stated":"explicit","quote":"We believe our Filters 1-3 are extremely conservative, and unlikely to classify organic behavior as wash trading."},{"assumption":"Only trades paid with ETH (or wETH) are included to ensure price consistency.","type":"data","stated":"explicit","quote":"the 0.39% of transactions not conducted in ETH-type currencies (e.g., ETH, wETH) to ensure consistency in our analysis."},{"assumption":"Roundness of trade price (last non‑zero digit >1% of price) indicates low wash‑trading activity and serves as a benchmark.","type":"model","stated":"explicit","quote":"Cong et al. (2023) categorize a trade price as “round” if its last nonzero digit exceeds 1% of the total price."},{"assumption":"Upstream Ether transfers are limited to external EOAs to conservatively capture potential collusion.","type":"data","stated":"explicit","quote":"we limited the transfers to be of “external” type, i.e., Ether transfers from Externally Owned Accounts (EOAs)."},{"assumption":"The threshold τ can be optimized to minimize aggregate prediction error.","type":"model","stated":"explicit","quote":"We refer to this approach as “τ-Regression” where the threshold τ is the outcome of an optimization."}]},"datasets":{"status":"available","items":[{"name":"LooksRare NFT sales","provider":"Alchemy getNFTSales API","kind":"crypto-onchain","access":"public","period":"11:04, December 29th, 2021 to 07:56, Apr","universe":"LooksRare marketplace transactions","quote":"For LooksRare, our data covers all transactions from 11:04, December 29th, 2021 to 07:56, April 13th, 2023."},{"name":"Blur NFT sales","provider":"Alchemy getNFTSales API","kind":"crypto-onchain","access":"public","period":"04:49, October 19th, 2022 to 17:14, Marc","universe":"Blur marketplace transactions","quote":"For Blur, our data covers all transactions from 04:49, October 19th, 2022 to 17:14, March 7th, 2024."},{"name":"OpenSea NFT sales","provider":"Alchemy getNFTSales API","kind":"crypto-onchain","access":"public","period":"19:57, February 18th, 2022 to 13:46, Mar","universe":"OpenSea marketplace transactions","quote":"For OpenSea, our data covers all transactions from 19:57, February 18th, 2022 to 13:46, March 20th, 2024."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"tick"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R4a","candidates":{"front_matter":"n/a","evidence_card":"partial","evidence_card_quoted":true},"card_detail":"Filters consider trading fees to the marketplace and Ethereum gas costs, but costs are not incorporated into headline error metrics.","card_quote":"Filters 1-3, only catch behavior where a user engages with the NFT marketplace (and pays trading fees to the marketplace on top of the Ethereum gas costs)","meaning":"the paper reports no trading or execution result; the cost the extractor found is a model parameter"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"Approximately 38% of NFT trades and 60% of traded value across three major exchanges likely involve wash‑trading.","kind":"statistical","quote":"we find that ∼38% (30–40%) of trades and ∼60% (25–95%) of traded value likely involve manipulation"},{"finding":"Error rates drop dramatically from 1‑Regression (45.6% on Blur) to τ‑Regression (31.0%) to ML models (0.4%).","kind":"performance","quote":"Table 6: Absolute Estimation Errors Across Platforms and Methods"}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"available","supports":[],"contradicts":[{"paper":{"paper_id":"2305.01543","title":"NFT Wash Trading Detection","paper_date":"2023-02-07","authors":["Derek Liu","Francesco Piccoli","Katie Chen","Adrina Tang","Victor Fang"],"empirical_rigor":4.5,"page_url":"https://thequant.space/flowcharts/nft-wash-trading-detection/"},"question":"What proportion of NFT trades involve wash trading?","this_paper_finding":"Approximately 38% of NFT trades and 60% of traded value across three major exchanges likely involve wash‑trading.","other_paper_finding":"On average 0.136% of transactions, 0.157% of tokens, and 0.109% of wallets are involved in wash trading, generating total price manipulation, sales, and repurchase profit of around $900K, $1.1M, and -$1.6M respectively.","other_paper_quote":"Over all collections that the algorithm was tested on, on average, 0.136% of transactions, 0.157% of tokens, and 0.109% of wallets in each collection are involved in wash trading (row 8, Table III-V). These wash sales generate an overall total price manipulation, sales, and repurchase profit of around $900K, $1.1M, and","note":"A finds 0.136% of transactions, B finds ~38% of trades; the estimates differ by orders of magnitude, reflecting different samples/metrics.","confidence":0.7},{"paper":{"paper_id":"2306.04643","title":"Abnormal Trading Detection in the NFT Market","paper_date":"2023-05-25","authors":["Mingxiao Song","Yunsong Liu","Agam Shah","Sudheer Chava"],"empirical_rigor":6,"page_url":"https://thequant.space/flowcharts/abnormal-trading-detection-in-the-nft-market/"},"question":"What proportion of NFT trades involve wash‑trading?","this_paper_finding":"Approximately 38% of NFT trades and 60% of traded value across three major exchanges likely involve wash‑trading.","other_paper_finding":"The wash trader percentage in the NFT market is estimated at 5.38% based on the clustering results.","other_paper_quote":"five of them are marked with specific labels while the other two are potential wash trader candidates, concluding that the wash trader percentage in the NFT market is 5.38%.","note":"A reports 5.38% wash traders; B reports ~38% of trades (and 60% of value) as wash‑trading, showing a large discrepancy.","confidence":0.666}],"qualifies":[],"compared_with":12,"citations":{"cited_by_in_corpus":0,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T18:18:28Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}