IP Insights
AI IN MOVIES: LEGAL SAGA
Estimated : 8 Mins
1. INTRODUCTION:
The film industry is undergoing a revolution with the integration of Artificial Intelligence (AI) technologies. AI is being used in various stages of filmmaking, from scriptwriting to post-production and audience engagement. In scriptwriting and content generation, AI-powered tools have the ability to assist in generating ideas, improving story structures, and even creating dialogue. However, there is a risk of unintentional similarities or plagiarism when using AI-generated content. AI models are trained in existing scripts and materials, which may lead to the inadvertent generation of content that resembles or infringes upon copyrighted works.
2. WHAT DOES “AI IN MOVIES” MEAN?
“AI” in modern filmmaking usually denotes generative models (large language models for text; image generators; neural voice models; video/temporal models; simulation/physics models) plus specialised tools (RVC voice cloning, upscales, denoisers, motion-capture enhancement, synthetic crowd generation).
Typical film uses:
- Development: rapid ideation, beat sheets, formatted screenplay drafts, logline testing.
- Pre-vis / concept art: instant concept frames, mood boards, storyboards.
- Production: virtual production assists (real-time compositing), background population, facial/performance tracking, and digital doubles.
- Post: de-noising, frame interpolation, upscaling, de-aging, synthetic dubbing/ADR, marketing assets (AI key art).
- Restoration & archives: automated colorisation and restoration of archival film.
During production and cinematography, AI is increasingly employed to enhance efficiency and quality. Autonomous drones equipped with AI algorithms capture complex aerial shots with precision, reducing the need for costly helicopters and skilled pilots. AI aids in camera tracking, motion capture, and optimizing camera settings for lighting conditions, resulting in better shots and reduced post-production workload. Post-production and visual effects (VFX) have undergone significant transformations through AI. Labor-intensive tasks are now automated, and creativity is enhanced. AI algorithms generate realistic special effects, manipulate footage, and seamlessly integrate CGI into live-action scenes. This accelerates the VFX process, allowing filmmakers to achieve stunning visuals more efficiently. AI technologies are instrumental in marketing and audience analysis. By analyzing data from social media platforms, online discussions, and audience feedback, AI provides valuable insights into audience preferences and trends. Additionally, AI-powered tools can assist in character development by analyzing vast amounts of data related to human behavior, psychology, and cultural references. This enables writers to create multi-dimensional characters with realistic traits and motivations. AI algorithms can offer suggestions for dialogue that aligns with the characters’ personalities, ensuring more authentic and engaging interactions. For example, Lokesh Kanagaraj in Coolie (2025): the director publicly acknowledged AI was used in aspects involving Rajinikanth’s material in Coolie and intends to use AI for Kaithi 2. Reports indicate use for voice assistance and production tasks (public interviews). This is an explicit high-profile Tamil example of creative teams adopting AI within mainstream production.
3. LEGAL ISSUES CREATED BY AI FOR MOVIES
3.1 Authorship & copyrightability: What is “original” human expression vs. machine-generated output? “Original” human expression arises from a creator’s skill, judgment, and creativity, while machine-generated output is algorithmic, however, when there is significant human intervention, such as selecting, curating, or creatively shaping the AI’s output, the resulting work may qualify as original. Many jurisdictions like UK and, USA, are concluding that human authorship is required for copyright registrations, leaving wholly autonomous AI output in a precarious position. Indian copyright law does not currently recognize non-human authorship, and AI-generated outputs without human “minimal creativity” fail registration/upload possibilities under the Copyright Act. According to Section 2(1)(d) “computer-generated works” that attribute authorship to the person who caused the work to be made. While that language is sometimes cited to argue for protection of some AI outputs, its fit to modern generative AI (large models producing novel images/text with minimal human “arrangement”) remains contested.
Thus, the usage of a human intervention is a key factor to claim copyright over such scenes/clips.
3.2 Training-data liability: Lawsuits assert that ingesting copyrighted images/text for model training can be infringing (direct/secondary theories). Plaintiffs seek damages and discovery into model training datasets and provenance. For example, Anil Kapoor won a landmark ruling in September 2024, preventing multiple defendants from using his likeness, voice or catchphrases (e.g., Jhakaas”) in deepfake/AI-generated media without consent for commercial or other use.
3.3 Performer’s Rights/voice rights: Actors’ voices and likenesses can be digitally replicated; unions and some states now require consent and compensation for digital replicas. Sections 38, 38A, and 38B of the Copyright Act, 1957 recognize performers’ rights for actors, singers, musicians, and dancers, giving them control over reproduction, communication, and distribution of their performances for 50 years For example, Singer Arijit Singh won a landmark case against unauthrozie voice cloning. For Indian cinema, this means that producers must obtain consent from living actors or legal heirs before deploying AI to recreate performances of stars such as M.G. Ramachandran or Sridevi.
3.4 Right of Publicity: Indian courts have also recognized personality rights under Article 21 of the Constitution and tort principles. In ICC Development (International) Ltd. v. Arvee Enterprises the Delhi High Court held that publicity rights exist independently of privacy rights. Similarly, in Titan Industries v. Ramkumar Jewellers (2012 (50) PTC 486 Del), the unauthorized use of Amitabh and Jaya Bachchan’s likeness in advertisements was restrained as a violation of their publicity rights. Earlier, in DM Entertainment v. Baby Gift House, Daler Mehndi successfully prevented unauthorized use of his persona. These precedents extend naturally to AI: a deepfake trailer or promotional clip using Rajinikanth’s likeness without permission would amount to misappropriation of his publicity rights, even if generated using AI tools.
3.5 Defamation, privacy & deepfakes: manipulated videos that impersonate living (or deceased) persons can be defamatory, privacy-invasive, or criminal under existing statutes (where applicable); India currently relies mainly on IT Act / IPC provisions while specific AI deepfake laws are limited.
3.6 Consumer protection & disclosure: EU-centric rules and proposed transparency regimes seek labeling of AI-generated content and may require publishing training-data summaries for certain models.
4. PRACTICAL COMPLIANCE CHECKLIST FOR INDIAN PRODUCTIONS
4.1 Contracting: include explicit AI-use clauses in all talent, writer, VFX and vendor agreements covering (a) permission for scans/voice modeling, (b) scope/medium/duration, (c) compensation for reuse, (d) no-training warranties where required, (e) indemnity for unauthorized inputs.
4.2 Data provenance and logs: maintain dataset inventories, prompt logs, version histories, and human-curation records (to prove human authorship or to trace training data). This helps in litigation and rights clearance.
4.3 Performer consent: obtain granular consents for digital replicas and voice synthesis; define reuse (sequels, trailers, ads) and preserve separate approvals for future uses. Follow SAG-AFTRA language when U.S. performers are involved.
4.4 Clear music & sound rights: verify whether AI-generated recreations of singers or songs implicate music publishing/labels; secure mechanical and moral-rights clearances where relevant.
4.5 Label AI-generated marketing where required: for EU/UK releases, track AI-disclosure laws and the evolving AI Act compliance needs.
4.6 Insurance & indemnities: seek vendor indemnities and update E&O (errors & omissions) policies to cover AI-driven IP exposures
CONCLUSION:
AI promises to transform Indian cinema, from streamlining production to enabling the digital resurrection of legendary stars. Yet, this transformation must operate within the boundaries of Indian copyright law, performers’ rights, and personality rights jurisprudence. While the Copyright Act, 1957, provides partial answers, courts are beginning to adopt existing law to address AI misuse. For Tamil and Indian filmmakers, the safest approach lies in obtaining explicit performer consent, licensing underlying works for AI training, and ensuring human creativity in AI-assisted outputs. Until Parliament enacts AI-specific legislation, Indian cinema will continue navigating these issues through judicial interpretation and contractual safeguards.
Popular Posts
