---
id: "framework-rightsholder-defense"
type: "framework"
source_timestamps: ["§ Lessons for Rightsholders"]
tags: ["business-strategy", "ip-protection", "monetization"]
related: ["action-curate-and-license", "action-implement-poisoning-tools", "action-rethink-freemium", "action-demand-retrain-removal"]
steps: ["Rethink Open Web Exposure: Evaluate the trade-offs of freemium/ad-supported models against the harms of unauthorized AI scraping. Consider moving valuable IP behind paywalls.", "\\\"Implement Technical Protections: Deploy robots.txt", "web-hosting scraper blocks", "and image-protection tools (Glaze", "Nightshade) for any content remaining on the open web.\\\"", "\\\"Curate and License: Package high-quality", "reliable datasets tailored for AI training and negotiate licensing deals with AI developers who need clean data quickly.\\\"", "Enforce Legal Rights: Join existing litigation or file new suits arguing that Gen AI threatens the livelihoods of the creators training the models.", "Demand Retraining Removal: Monitor major LLM version updates and demand the extraction of copyrighted material during the from-scratch retraining phase."]
sources: ["tail2"]
sourceVaultSlug: "hbr-seg-tail2"
originDay: 2
articleStem: "hbr-tail-126-genai-copyright"
sourceUrl: "https://hbr.org/2025/07/can-gen-ai-and-copyright-coexist"
sourceTitle: "Can Gen AI and Copyright Coexist?"
---
# Rightsholder AI Defense Strategy

A comprehensive, multi-pronged defense strategy for IP rightsholders in the generative-AI era, from [[entity-michael-d-smith]] and [[entity-rahul-telang]]. Because courts are divided on fair use (see [[concept-fair-use-divergence]]), rightsholders cannot rely on litigation alone; they must combine technical defense with proactive monetization.

**The five moves:**
1. **Rethink open-web exposure** — weigh freemium/ad-supported models against scraping harms; move valuable IP behind paywalls. → [[action-rethink-freemium]], evidence in [[claim-paywall-protection]].
2. **Implement technical protections** — robots.txt, host-level scraper blocks, and image-protection tools like [[entity-glaze-nightshade]] for content that must stay on the open web. → [[action-implement-poisoning-tools]].
3. **Curate and license** — package clean, reliable datasets for AI training and sell licenses to developers who need risk-free data fast. → [[action-curate-and-license]], concept [[concept-curated-training-datasets]].
4. **Enforce legal rights** — join or file suits arguing that Gen AI threatens the livelihoods of the very creators who produced the training data.
5. **Demand retraining removal** — monitor major model version transitions (e.g., GPT-3 → GPT-4) and demand removal of your works during the from-scratch retrain. → [[action-demand-retrain-removal]], mechanism [[concept-model-retraining-removal]].

The unifying logic: technical defense (paywalls, robots.txt, Glaze/Nightshade) plus proactive monetization (curated licensing) plus lifecycle-timed legal pressure. This is the rightsholder-side mirror of [[framework-gen-ai-risk-mitigation]].