AI Tokens as an Emerging Asset Class: Data and Methodology Dr Yuqian Zhang, Auckland University of Technology Date: 23 July 2026 ================================================================================ OVERVIEW ================================================================================ This directory contains the datasets used in the research brief "AI Tokens as an Emerging Asset Class: Economic Logic, Market Evidence, and Accounting Implications." Each dataset is described below with its source, compilation method, and known limitations. ================================================================================ DATASET DESCRIPTIONS ================================================================================ 1. ai_tokens_market_cap.csv Description: Quarterly AI token sector market capitalisation and dominance relative to total crypto market (2020 Q1 to 2026 Q2). Sources: Compiled from CoinGecko, CoinMarketCap AI & Big Data category aggregates, and individual token data. Early-period estimates (2020-2021) are approximate due to limited AI token categorisation at the time. Key columns: year, quarter, total_market_cap_usd_billion, btc_market_cap_usd_billion, ai_token_count, ai_token_dominance_pct Limitations: Market cap data are end-of-quarter snapshots and do not capture intra-quarter volatility. AI token category definitions vary across data platforms. Tokens listed as AI-related include some borderline cases (e.g., data storage tokens used by AI projects). 2. ai_tokens_top_projects.csv Description: Major AI-focused blockchain projects by market capitalisation as of Q2 2026. Sources: CoinMarketCap AI & Big Data category (July 2026 snapshot), individual project documentation, CoinGecko. Key columns: token, ticker, category, launch_date, market_cap_usd_billion_q2_2026, ath_price_usd, ath_date, use_case, blockchain Limitations: ATH figures from CoinMarketCap may differ from exchange- specific data due to price aggregation methodology. Market cap rankings are point-in-time and shift rapidly in this sector. 3. ai_tokens_price_history.csv Description: End-of-quarter price and market cap for five major AI tokens (TAO, RNDR, FET, AKT, NEAR) from Q1 2023 to Q2 2026. Sources: CoinMarketCap and CoinGecko historical data endpoints. Key columns: token_ticker, token_name, year, month, price_usd, market_cap_usd_million Limitations: Prices are approximate quarter-end values. FET data before the ASI merger (July 2024) refers to the original FET token; post-merger reflects the Artificial Superintelligence Alliance token. 4. ai_tokens_corporate_adoption.csv Description: Corporate holdings, treasury strategies, and institutional investment vehicles related to AI tokens and digital assets. Sources: SEC EDGAR filings, corporate press releases, Bitcoin Treasuries (bitcointreasuries.net), Grayscale product pages. Key columns: company, type, holding, estimated_value_usd_million, date, notes Limitations: Corporate AI token holdings are difficult to verify comprehensively. Most publicly listed firms do not break out AI tokens separately from general crypto holdings. Grayscale fund AUM is not publicly disclosed in real time. 5. ai_tokens_vc_funding.csv Description: Quarterly venture capital funding in the AI-crypto sector. Sources: Compiled from PitchBook, Crunchbase, Messari, and project press releases. Aggregate figures are estimates based on available deal data. Key columns: year, quarter, vc_funding_usd_billion, number_of_deals, notable_deals Limitations: Many crypto VC deals are undisclosed or announced without funding amounts. Data likely undercounts seed-stage and private investments. Figures are estimated aggregates and may differ from proprietary databases. 6. ai_tokens_accounting_comparison.csv Description: Comparison of accounting frameworks applicable to AI tokens under US GAAP and IFRS. Sources: FASB ASU 2023-08, IAS 38, IAS 2, IFRS IC Agenda Decision (June 2019), SEC SAB 121, EU MiCA. Key columns: standard, framework, classification, measurement, effective_date, key_features Limitations: The IASB has not issued a dedicated crypto-asset standard. IFRS classification depends on facts and circumstances of each holder. This table represents a general framework and may not apply to all specific arrangements. 7. ai_tokens_regulatory_timeline.csv Description: Chronology of major regulatory and standard-setting developments affecting AI tokens and crypto assets. Sources: SEC, FASB, IASB, European Commission, BIS, official press releases and publications. Key columns: date, event, jurisdiction, category, significance Limitations: Focuses on major jurisdictions (US, EU). Does not include all national-level regulatory actions. Timelines for proposed rules may shift. 8. ai_tokens_academic_literature.csv Description: Selected academic articles from ABS 4/4* and ABDC-A* journals relevant to AI tokens, crypto assets, and tokenisation. Sources: Scopus, Web of Science, SSRN, journal websites. Key columns: journal, title, authors, volume, pages, year, topic, evidence_type, key_findings Limitations: Not a systematic review. Selection emphasises top-tier accounting and finance journals. Literature on AI tokens specifically is nascent; many entries are on broader crypto/blockchain topics. ================================================================================ DATA QUALITY AND REPRODUCIBILITY ================================================================================ All datasets were compiled manually from public sources and cross-checked where possible. Data were verified against at least two independent sources for major figures (market capitalisation, regulatory dates, corporate holdings). Remaining discrepancies are noted in the limitations for each dataset. The Python replication script (scripts/replicate.py) loads these datasets and reproduces all charts and statistics in the report. For data obtained from public, open-access sources, the script auto-downloads fresh data where available. For manually compiled data, the script reads directly from CSV files. Last updated: 23 July 2026