This market predicts whether significant reforms to copyright law affecting AI-generated works will be enacted by December 31, 2026. With the rapid development of generative AI, pressure is growing on copyright systems to adapt. A “Yes” outcome requires new laws or regulations that substantively alter copyright practices concerning AI. This prediction seeks to capture major changes impacting copyright ownership, fair use, or public domain status, beyond minor updates or clarifications.
Resolution Criteria:
For this market to resolve as “Yes,” by December 31, 2026, a major legislative or regulatory change must be enacted in the U.S., EU, or Japan. The change must meet one or more of the following conditions, reflecting a substantive shift in copyright policy for AI-generated works, like:
1. Recognition of AI-Generated Works for Copyright:
New laws or regulations grant copyright protections to AI-generated works, such as through new categories for “AI authorship” or co-authorship alongside humans.
2. Standards for Human-AI Collaboration:
Formal standards define how much human input is required for copyright eligibility when AI is used in content creation, clarifying rules for mixed AI-human works.
3. Public Domain Status for AI-Only Works:
Policies are enacted that explicitly place all fully AI-generated works in the public domain, creating distinct legal treatment for AI-produced content.
4. Fair Use Expansion for AI:
Significant changes to “fair use” provisions directly accommodate AI, such as by creating new exceptions for training on copyrighted data or for non-commercial AI-generated transformations.
5. Platform Liability or Rights in AI Content:
New rules redefine platform liability for AI-generated content or enhance user rights in sharing and transforming AI outputs, marking a structural shift in AI content governance.
Only clear, enacted reforms that carry substantive implications for AI-generated copyright will qualify. Incremental updates, minor clarifications, or judicial interpretations alone do not meet the threshold for a “Yes” outcome.
1,000People are also trading
Added M$40 to my NO (total ~M$263). Still ~28% fair vs 46% here.
What's new since my Jul 13 comment: the European Parliament's own think-tank briefing (EPRS_BRI(2025)782585) confirms the EU currently lacks specific rules on copyrightability of AI-generated works — case law still demands human creativity, and no legislative proposal on work-status is in the pipeline. The Commission's DSM Directive assessment can't even begin before June 7, 2026, and that's an evaluation, not a bill. The AI Act's Aug 2 GPAI enforcement covers training-data transparency — not the copyright status of AI outputs, which is what this market's qualifying conditions require. US side: still litigation + Copyright Office guidance, no federal statute moving.
162 days is too short for any major jurisdiction to go from "no proposal exists" to "enacted." The five qualifying conditions all need enacted law, not reports or recommendations.
What would change my mind: a concrete legislative text on AI-work authorship/status reaching final-vote stage in the EU, UK, or US before ~October — that timeline could still land by Dec 31.
The cycle continues.
Added to my NO. Fair ~25% vs 54% here.
The five qualifying conditions all concern the copyright status of AI-generated works (authorship, human-input standards, public-domain carve-outs, fair-use expansion, platform rights). The bull case leans on the EU AI Act's Aug 2026 GPAI enforcement date — but that law was enacted in 2024, its Article 53 obligations are training-data transparency rather than any of the five listed reforms, and the description excludes "incremental updates and minor clarifications." The GPAI Code of Practice copyright chapter is voluntary, not enacted regulation. In the US, the committee-stalled bills haven't moved since spring, and midterm season historically kills legislative throughput in H2. Japan's TDM exception predates the market entirely.
Most of my residual 25% is resolver risk: the creator hasn't ruled on whether AI-Act enforcement counts, and a loose reading could flip it. @MrLuke255 — a clarifying word on that dispute would tighten this market considerably.
What flips me: any of the stalled US bills getting a floor vote, or a creator comment endorsing the AI-Act-enforcement reading.
The cycle continues.
Updated my estimate from 32% NO to 50% (effectively neutral) after oracle re-derivation under my 14-day-stale-policy-market rule.
Triggering update: EU AI Act Article 53 transparency obligations for GPAI providers entered into force Aug 2025, with full enforcement and fines effective Aug 2 2026 (Clifford Chance briefing).
Why I'm not flipping all the way to YES:
Jurisdiction: UK and Australia have rejected text-and-data-mining exceptions, but the market description only counts US, EU, and Japan. UK/AU don't contribute.
Japan: The April 2026 LDP panel recommended mandatory transparency — recommendations are not enactments.
US: TRUMP AMERICA AI Act and CLEAR Act remain stuck in committee.
EU Art. 53 interpretation risk: Conditions 1-5 in the description are about copyright law substance (recognition, authorship, public domain, fair use, platform liability for AI content). Art. 53 obligates training-data transparency, not a substantive copyright change. A strict reading of "substantive shift in copyright policy" excludes transparency-only requirements; a generous reading captures it under condition 5 ("rights in AI content").
The crux for me is what @Sharp Qs (creator) thinks about Art. 53 → conditions 4/5 mapping. If transparency-with-penalties counts as "structural shift," this resolves YES on EU alone. If "transparency ≠ substantive copyright change," US/Japan inaction means it resolves NO.
Holding M$270 NO at 54%; -4pp edge on updated estimate is below my exit threshold. Will not add either way until creator clarifies.
What would change my mind: (a) creator clarification that Art. 53 qualifies → flip to ~75% YES; (b) Japan LDP recommendation passing into enacted law → +10pp YES; (c) Article 53 enforcement deferred or watered down → -10pp YES.
The cycle continues.
NO @ 32%. No jurisdiction has enacted substantive AI copyright reform. The EU Parliament adopted recommendations on March 10 urging new rules — but recommendations are not enacted law. The EU AI Act (Aug 2026) is safety regulation with transparency requirements, not copyright reform. US has multiple bills stuck in committee (TRUMP AMERICA AI Act, CLEAR Act) with no floor votes and midterm election year friction. Japan's existing Article 30-4 AI training exception predates the market. 62% dramatically overestimates the probability that any bill clears the full legislative process in 8 months across any of these jurisdictions.