OpenAI’s Sora app from ChatGPT maker tests limits of copyright

OpenAI's new Sora 2 app has rapidly become a leading video generation tool, demonstrating advanced capabilities in creating high-fidelity video from text prompts, including lifelike cameos and synchronized audio. However, its swift rise is immediately overshadowed by significant legal and ethical concerns, primarily stemming from widespread user-generated content infringing on copyrighted material and the ease of creating sophisticated deepfakes. Legal experts criticize OpenAI's 'opt-out' approach for copyright as legally untenable, signaling substantial litigation risk for the company and the broader AI industry, as evidenced by Anthropic's recent $1.5 billion settlement, highlighting escalating intellectual property challenges and potential regulatory pressures within the rapidly advancing AI media creation sector.

Analysis

OpenAI's new Sora 2 text-to-video application has demonstrated significant technological advancement and achieved immediate viral adoption, securing the top position in its iOS App Store category. However, this launch is fundamentally overshadowed by substantial legal and ethical risks that pose a material threat to its commercial viability. The primary issue stems from the generation of outputs containing copyrighted material, such as characters from Nintendo and depictions from other media properties, which legal experts argue is indefensible under OpenAI's proposed 'opt-out' policy. This legal exposure is not theoretical; competitor Anthropic's recent $1.5 billion settlement for similar copyright infringement claims provides a tangible benchmark for potential damages. The controversy extends beyond OpenAI, implicating the broader generative AI sector, including competitors like Google, and creating uncertainty for major IP holders such as Disney, Warner Bros., and Sony. Furthermore, the platform's capacity for creating sophisticated deepfakes, as illustrated by a fabricated video of CEO Sam Altman, raises significant reputational and societal risks. While OpenAI has implemented mitigation techniques like watermarking, its own documentation admits these are not 'a silver bullet' and can be easily removed, rendering them ineffective against intentional misuse.

Key Decisions for Investors

• Investors in media conglomerates holding significant intellectual property, such as Disney (DIS) and Sony (SONY), should monitor for litigation announcements, as the widespread infringement by Sora 2 presents a strong case for pursuing substantial damage claims, following the precedent set by Anthropic's $1.5 billion settlement.
• The legal and ethical issues surrounding Sora 2 should be viewed as a systemic risk for the entire generative AI sector, prompting investors in companies like Alphabet (GOOGL) to re-evaluate legal risk premiums and scrutinize data sourcing and output policies across their AI investments.
• For investors with any direct or indirect exposure to OpenAI, the platform's immediate viral success must be heavily discounted by the immense legal liability from its 'move fast and break things' approach to copyright, making potential lawsuits from major IP holders the single most critical factor to watch for determining long-term value.