Search papers, labs, and topics across Lattice.
This paper critiques existing market designs for human-generated content used in AI training, highlighting the inadequacies of both free-for-all and strong intellectual property rights models. By modeling these scenarios as a static Stackelberg game, the authors reveal that both approaches fail to adequately incentivize creators, particularly innovative ones, leading to what they term the "originality penalty." They propose a dynamic market design that incorporates a data intermediary to address cross-creator externalities and enhance the quality of contributions, ultimately improving AI model performance.
The "curse of precision" reveals how reliance on AI-generated content can degrade model performance by homogenizing training data.
How can we design a market of human-generated content for use in training AI models that both enables technological progress and preserves individual incentives for high-quality content creation? Existing approaches take polar positions: a "free-for-all" model based on fair use and a "strong intellectual property rights" model. We show that both fail: Free-for-all does not compensate creators, and -- by modeling as a static Stackelberg game -- strong intellectual property rights also underpower creative incentives. We find this especially true for more innovative creators, a phenomenon we term the "originality penalty." Extending this insight to a dynamic model, we find another market failure undermining AI model performance, even for an initially good model: Such a model induces greater reliance by humans on AI-assisted creation, resulting in homogenized content feeding back into training, which degrades the model performance -- a "curse of precision." We further propose a market design with a data intermediary internalizing cross-creator externalities and subsidizing innovative contributions, thereby restoring efficiency.
MIT CSAIL