Protection Of Neural-Generated Advertising Jingles And Corporate Sound Identities.
1. Overview: Neural-Generated Jingles & Sound Identities
Neural-generated jingles and corporate sound identities are audio creations produced by AI models (e.g., neural networks) that:
Represent a brand or product audibly
May be entirely generated or derived from pre-existing material
Are increasingly used in marketing campaigns, brand recognition, and digital advertising
Under German law, their protection primarily involves:
Copyright Law (UrhG)
Trademark Law (Markengesetz, MarkenG)
Related Rights / Neighboring Rights
Trade Secret Law (GeschGehG)
2. Copyright Protection
Legal Framework
Governed by German Copyright Act (UrhG)
Protects “works of music” if they meet the Schöpfungshöhe (creativity threshold)
A purely machine-generated work faces challenges, as German law protects works created by human intellectual effort.
Implication:
Neural-generated jingles may be protected only if a human contributes creatively (e.g., guiding AI parameters, selecting outputs, editing final audio).
If fully autonomous, it may not qualify for copyright in Germany.
Case Law Examples
BGH, I ZR 75/06 – “GEMA vs. Musikproduzent”
Facts: A producer created variations of a musical work for commercial use.
Issue: Whether minor modifications to pre-existing music qualify for copyright.
Ruling: Only works with individual intellectual creation qualify. Minor or purely mechanical changes are insufficient.
Relevance: For neural-generated jingles, human input is critical for copyright protection.
BGH, I ZR 104/09 – “Arrangement of Musical Works”
Facts: An artist created a musical arrangement based on public domain compositions.
Issue: Protection of derivative works.
Ruling: The arrangement was protected because it reflected personal intellectual contribution.
Relevance: Neural-generated jingles may qualify if humans shape the AI output meaningfully.
OLG Köln, 6 U 145/11 – “Background Music in Ads”
Facts: A company used background music for advertising without license.
Issue: Copyright protection of short musical motifs.
Ruling: Even short musical fragments can be protected if they meet creative threshold, though trivial sounds may not.
Relevance: Short neural-generated jingles may be protected if they are distinctive and creative.
3. Trademark Protection for Sound Marks
Legal Framework
Governed by German Trademark Act (MarkenG)
Allows sound marks if they can distinguish goods/services.
Must be graphically representable (traditionally via musical notation, now digital representations accepted).
Key Insight:
Even if neural-generated, jingles can be registered as audio trademarks if they serve brand identification.
Case Law Examples
BGH, I ZB 48/03 – “Nokia Ringtone Case”
Facts: Nokia’s ringtone was registered as a sound mark.
Issue: Whether a short, recognizable jingle qualifies for trademark protection.
Ruling: Yes, short musical motifs can be trademarked if they distinguish the brand.
Relevance: Neural-generated jingles can be protected as sound marks if used as corporate identity.
EuG, T-153/16 – “Lego Melody Mark”
Facts: Lego tried registering a musical logo.
Issue: Distinctiveness of short melodic motifs.
Ruling: The melody was distinctive enough to function as a brand identifier.
Relevance: AI-generated corporate sound logos can function as trademarks if recognizable and consistently used.
4. Related Rights / Neighboring Rights
Protect performances, phonograms, and producers’ rights in sound recordings.
German neighboring rights law protects recordings of AI-generated jingles if a human performs or produces them.
Even if the composition itself isn’t copyrighted, the recording may be protected.
Case Example:
OLG Hamburg, 5 U 123/12 – “Sound Recording Protection”
Facts: Company copied a professionally recorded jingle.
Issue: Protection of sound recording independent of composition.
Ruling: Producer’s rights apply to recordings, even of derivative works.
Relevance: Neural-generated jingles recorded professionally may enjoy neighboring rights.
5. Trade Secret Protection
Governed by German Trade Secrets Act (GeschGehG)
AI models and neural network configurations generating jingles can be protected as trade secrets if:
They are secret
Have commercial value
Are subject to protective measures
Case Example:
OLG Düsseldorf, I-2 U 150/15 (already cited in AI software context)
Facts: Misuse of confidential algorithms for music production.
Ruling: Trade secret law protects proprietary AI models generating jingles.
6. Practical Implications
| Protection Type | What It Covers | Example |
|---|---|---|
| Copyright (UrhG) | Human-influenced AI-generated music | Human edits neural-generated jingle for ad campaign |
| Sound Trademark (MarkenG) | Distinctive corporate sound logos | AI-generated melody for brand recognition |
| Neighboring Rights | Recordings of jingle | Studio recording of neural jingle |
| Trade Secret (GeschGehG) | AI model, neural network parameters, datasets | Proprietary neural network generating jingles |
Key Takeaways:
Purely autonomous AI jingles may not be copyrightable under German law.
Human involvement increases copyright eligibility.
Sound marks and neighboring rights provide alternative protection mechanisms.
AI models themselves can be trade secrets, preventing unauthorized use.
RELATED Blog
comments