Introduction
Unlike the old days, when video games came on CD-ROMs and cartridges, modern video games are no longer standalone products. Instead, they depend on servers, authentication tools, and ongoing support just to function. Many modern video games today are even referred to as “live service” games. When support ends, live service games can no longer function as intended. The practice of ending support for live service games, known as “sunsetting,” has drawn increasing attention, including from campaigns like “Stop Killing Games” and other similar movements in Europe.
California’s proposed AB 1921[1], the Protect Our Games Act, responds directly to this discussion. For certain digital games first sold on or after January 1, 2027, the bill would require operators to provide advance notice before discontinuing the services necessary for a game’s “ordinary use” and, when those services end, ensure the game remains usable or provide a refund.
While the spirit of the “Stop Killing Games” initiative, and AB 1921, may be noble, practical implementation of the law poses a worrying challenge for publishers and developers.[2]
The statute defines “ordinary use” as access to a game’s core features based on how the game was marketed. First, for games that require online services to function, such online services become necessary to maintain “ordinary use.” From the perspective of publishers and developers, maintenance of an eternal online service can become an untenable burden. Second, AB 1921 does not account for third-party license dependencies. Finally, while AB 1921 acknowledges that there should be some exclusions, the manner in which it carves out such exclusions does not map onto the reality of how video games work. The European Commission has recently reached similar conclusions with respect to the “Stop Killing Games” initiative.[3]
Owned vs Licensed
Before addressing the substantive issues with AB 1921, we wish to address somewhat of a red herring in this whole discussion: whether video games are “owned” or “licensed.” The reality is that since the dawn of software products, video games—which are software products—have always been “licensed.” Consumers intuitively know this, because that is how things have always worked with respect to software. What the “Stop Killing Games” initiative is really trying to do is force video game companies to treat post-sunset live service games the same as traditional, single-player, offline premium games. For the reasons discussed herein, this is just not feasible.
1. The Limits of Live Service Games
AB 1921 incorrectly assumes that “ordinary use” of live service games can be preserved without imposing radical and extreme burdens on video game companies. Consider Destiny 2, a live service game that has recently been in the news because Bungie has decided to cease ongoing development (though Bungie is continuing to make the game available to players for the foreseeable future). Destiny 2 is fundamentally a multiplayer game, meaning that if AB 1921 were in effect, and Bungie chose to end all support for Destiny 2, Bungie would have to figure out a way to make it possible for end users of the game to continue with “ordinary use” of the game—that is, to continue enjoying multiplayer functionality.
In order to do so, Bungie would not only have to cease all online-authentication of users of Destiny 2, it would also have to perform at least the following tasks:
- Transfer all account information to each end user, of which there are millions.
- Release proprietary server-side software to enable end users to spin up their own servers, or, alternatively, undergo significant development efforts to patch the live service game to function in commercially available private server environments (which may be technologically impossible without rebuilding the game, in some cases).
- Release proprietary trade secrets with respect to matchmaking methodologies for PvP environments.
- Modify the game to remove gameplay dependencies on online services.
- Running QA cycles to verify the usability of the newly created offline version of the game.
To the extent that AB 1921 would necessitate release and disclosure of proprietary intellectual property which end users did not even pay for, it raises many legal concerns, including constitutional ones.[4] To the extent that compliance with AB 1921 can be met by simply re-building a live service game to function either entirely offline or within commercially available private server environments, such a law would force upon video game companies tremendous expenditures—upwards of the cost of developing a video game port—for free.
2. Third-Party Licensing and Contractual Conflicts
AB 1921 does not account for third-party rights. Video games often include licensed third-party content such as music, likeness rights, or other protected content that typically have fixed terms. When those terms expire, the operator may no longer have the right to distribute or maintain those elements. Removing that content is not always a workable solution; at best, it would require significant development efforts to replace all of the removed licensed content.
There is also the consideration of outbound license from a video game developer to a third party, such as a film studio. Those license agreements carry guarantees about the game—guarantees which may be violated by compliance with AB 1921.
3. Ambiguity in the Statutory Exceptions
Section 20664(b) of AB 1921 attempts to limit the statute’s reach by carving out three categories of video games that are exempt from compliance: (1) subscription-only services; (2) free-to-play games; and (3) games that are already permanently available offline.
In practice, these categories do not align cleanly with how games are actually distributed since many games exist somewhere in between:
A game may be sold for a one-time fee but still require constant server access for core functionality such that sunsetting it would “kill” the game (Arc Raiders).
A “free-to-play” game may generate substantial revenue through sales of content in the form of DLCs, battle passes, or other microtransactions (Sims 4).
A subscription-based game may also sell expansions or standalone content that require one-time fees commensurate with premium games (World of Warcraft).
A game may have previously required a fee but then later transition to a “free-to-play” model where older content is made available for free, but later content requires a premium purchase (Destiny 2).
The result is a framework that attempts to draw bright lines between business models that, in reality, overlap, defeating the purpose of the law. Section 20664(b) demonstrates that the drafters of AB 1921 have a fundamental misunderstanding of the video game industry. What AB 1921 would do is incentivize all developers to shift their business model to one of the excluded categories—something many gamers would lament.
4. EU Commission
On June 16, 2026, the European Commission published its legal and political conclusions on the “Stop Killing Games” initiative. The Commission ultimately agreed to explore with the video game industry the possibility of a self-regulatory code of conduct with respect to sunsetting games, but otherwise refused to impose legislation mandating the demands of the “Stop Killing Games” initiative. In reaching its conclusion, it balanced the intellectual property rights of the video games’ creators, the coverage of existing consumer protection laws, and the commercial burdens that would be imposed upon the video game industry.[5]
Conclusion
AB 1921 starts from an understandable concern: consumers should not be left with unusable products when support for a digital game ends. But as drafted, the bill treats live service games as though they can be preserved like traditional offline titles, without accounting for server architecture, third-party licenses, intellectual property constraints, or the blended business models that define the modern games industry. A workable solution should focus on clear disclosures, realistic sunset notices, and industry-specific standards that protect consumers without forcing developers to rebuild, disclose, or indefinitely maintain products in ways that were never technically or commercially contemplated. Until then, AB 1921 remains less a practical preservation measure than a reminder that consumer protection rules for games must be written with the realities of game development in mind.
[1] A.B. 1921, 2025–2026 Reg. Sess. (Cal. 2026).
[2] While there are also objections grounded in U.S. copyright law, the purpose of this blog article is to focus on the practicality and workability of AB 1921.
[3] Commission Communication, Communication from the Commission on the European Citizens’ Initiative (ECI) ‘Stop Destroying Videogames’, C (2026) 4110 final (June 16, 2026), https://citizens-initiative.europa.eu/document/download/75d642bc-6ff5-4713-b1cf-14f4aaf15869_en?filename=C_2026_4110_EN.pdf.
[4] U.S. Const. amend. V.
[5]Commission Communication, C (2026) 4110 final, at 3–6.

/Passle/644c41cc474c4c94b77327c8/SearchServiceImages/2026-06-19-22-49-54-737-6a35c792fe67db3561942de7.jpg)

/Passle/644c41cc474c4c94b77327c8/SearchServiceImages/2026-06-15-19-08-21-444-6a304da5824a95fe488994b3.jpg)
/Passle/644c41cc474c4c94b77327c8/SearchServiceImages/2026-06-15-18-36-50-989-6a304642705a1110f2c72d47.jpg)