According to @DeepLearningAI, websites are increasingly countering AI crawlers with decoys, blocking rules, and paywalls, marking an escalation in the online data access battle often framed as a shadow war. Source: DeepLearning.AI tweet, Nov 1, 2025. Evidence of tightening data supply includes The New York Times filing a copyright lawsuit against OpenAI and Microsoft over training on news content without permission. Source: The New York Times, Dec 27, 2023. Platforms are monetizing access via licensing, highlighted by Reddit’s partnership with Google to provide real-time content for model training and enhanced search. Source: Reddit press release, Feb 22, 2024. Developer knowledge platforms are also striking paid data deals, such as Stack Overflow’s partnership with OpenAI to integrate Stack Overflow data into model training and services. Source: Stack Overflow press release, May 6, 2024. For traders, constrained open-web data and the shift to licensed pipelines signal rising AI training and inference input costs and value accrual to rights holders and data vendors. Source: Reddit press release, Feb 22, 2024; Stack Overflow press release, May 6, 2024. In crypto, decentralized data and compute projects positioned around licensed access and provenance are relevant monitoring candidates, including Ocean Protocol (OCEAN) for tokenized data marketplaces, Render Network (RNDR) for decentralized GPU compute, and Fetch.ai (FET) for agentic AI networks. Source: Ocean Protocol documentation; Render Network documentation; Fetch.ai documentation. A key operational risk is broader adoption of robots.txt blocks against AI crawlers like GPTBot, which formalizes compliance-driven access and pushes model builders toward paid feeds. Source: OpenAI blog announcing GPTBot and crawler-blocking via robots.txt, Aug 7, 2023.