A philosophical and legal analysis of 'spontaneous data' — human-generated content like Instagram photos and TikTok videos — and its use in training AI systems to produce sexually explicit content. The piece covers the Software 2.0 paradigm where data rather than code is central, explains why existing intellectual property frameworks (Lockean labor theory and utilitarian incentive theory) fail to protect spontaneous data creators, and examines why consent alone is insufficient protection. Three counterarguments are explored: private use, anonymized transformation, and a utilitarian case for AI-generated explicit content as a harm-reduction mechanism. The article also touches on deepfakes, the interpretability problem in AI, and the philosophical distinction between human-level and metaphysically-aligned cognition.
