Training artificial intelligence models with works on the internet is the subject of copyright infringement lawsuits around the world.
Notable Cases Filed
- Getty Images vs Stability AI
- NYT vs OpenAI
- Authors group vs OpenAI
- Artists vs Midjourney
Controversy Points
- Education data use = duplication?
- Does the "Fair use" / "fair dealing" exception apply?
- Is the output a work created by humans or AI?
- The author's right to compensation?
Situation in Türkiye
- No precedent case yet
- FSEK general provisions apply
- License for commercial use may be required
Supreme Court/EU Approach
- EU AI Act: "transparency" requirement in training data
- Manufacturers must publish a list of training data
- Owners' right to opt-out blocking
- Requesting a license
- Compensation (still controversial)
Responsibilities of Companies
- Training data list transparency
- License with authors
- Opt-out mechanisms
- Compensation in case of violation
Practical Advice (Author)
- On your website Add AI bot access barriers to robots.txt (GPTBot, ClaudeBot)
- Record questionable AI output
- Consider joining class action lawsuits
- Seek legal advice
Intellectual property and technology lawyer recommended.