The namesake of the debate, Benoit Mandelbrot, is best known as the founder of fractal geometry, a field of mathematics that deals with the study of infinitely repeating geometric shapes. Over the course of his career, Mandelbrot developed numerous algorithms that can be used to calculate these fractals. An algorithm is a precisely defined sequence of steps that are used to solve a specific task – in this case, to generate a fractal.
In the 1970s, the first discussion arose as to whether the fractals used by Mandelbrot as the result of a calculation sequence could be protected by copyright. Mandelbrot argued that his algorithms had the necessary creativity and level of creation and should therefore also be protected by copyright.
The other side argued that algorithms are merely mechanical processes and therefore not creative works within the meaning of copyright law.
A final solution to the conflict has not yet been found, as the arguments on both sides still appear justified.
The legal situation in Germany
In Germany, algorithms are sometimes regarded as “creative works” and are then protected by copyright. This means that the author of an algorithm has the exclusive right to use and distribute the algorithm. However, there are also some restrictions here. The degree of creativity required of the author is difficult to measure and the threshold for achieving the copyright protection standard is quite high.
For example, algorithms that merely represent an “ideal-type” solution to a problem and do not contain any specific instructions on how the problem can actually be solved are not eligible for protection. Algorithms that are based on already known rules or procedures are also not eligible for protection, as they do not constitute an “individual intellectual creation”. The actual scope of copyright protection must therefore be determined for each individual case; there is no general answer to this question. In most cases, however, it is assumed that the necessary level of creation has not been reached.
An important difference between algorithms and other types of intellectual property, such as pieces of music or novels, is that algorithms can usually be used repeatedly to produce new results, whereas most other types of intellectual property can only be used once. This makes it difficult to control the use of algorithms and determine their monetary value. An author’s income can in principle be protected by appropriate licensing, but what determines the “reasonableness” of a license fee for algorithms is unclear.
Protection through secrecy
It is also problematic that the algorithm would have to be disclosed in the context of copyright infringement proceedings. As a rule, algorithms from the corporate context cannot be protected under the Copyright Act, but can be protected under the Trade Secrets Act. Depending on the use case, however, this protection of secrets would then be lost in a lawsuit. Competing companies could acquire the specialist knowledge disclosed in this way and profit from it without any further protection mechanism being provided for under German law.
Such a system would virtually provoke abuse, as the infringer could circumvent the competitor’s internal confidentiality rules as part of the process.
Debate with a future
The debate on the copyright protectability of algorithms in Germany shows how difficult it is to apply the concept of intellectual property in the digital world. Although algorithms can in principle be creative works, there are numerous restrictions and exceptions. The threshold for protection under copyright law is high, and the protection mechanisms of the Copyright Act do not apply to most algorithms.
It remains to be seen how the legal situation regarding algorithms will develop in the future. Reconciling the requirements of the digital world with the rights of algorithm creators is likely to be the biggest hurdle. represent the biggest hurdle. At the same time, the use of algorithms must remain practicable for society.
Particularly in times of increasing “intelligence” of artificial intelligence, this debate is becoming increasingly relevant, provided that no clear legal regulation has been found. For the copyright protection of computer programs – which are the result of numerous algorithms – Section 69a UrhG has explicitly stated since 1993: ” Computer programs are protected if they constitute individual works in the sense that they are the result of their author’s own intellectual creation. No other criteria, in particular qualitative or aesthetic criteria, are to be applied to determine their protectability “.
BGH: Avoid monopolization of ideas
The BGH has so far refused to include new absolute copyrights. Software companies such as IBM and Microsoft have already attempted to extend the scope of protection of the Copyright Act in the IT sector in this way, but have failed due to the attitude of the BGH (BGH, judgments of 24.02.2000 – I ZR 141/97 and of 06.07.2000 – I ZR 244/97).
The central aspect of the Federal Supreme Court’s reasoning is that a monopolization of ideas must be avoided at all costs. Technological progress would be significantly slowed down by excessive copyright protection for algorithms or simple programs.
It remains to be seen to what extent the BGH will change its position in the future. However, it is clear that the problem will not go away on its own. The more technical progress is made in connection with computer-generated results, the greater the need for legal protection for the respective “authors”.
Focus on AI & copyright
In our thematic focus “AI & Copyright“, we deal with the copyright aspects of current developments in the field of artificial intelligence. So far, this series has included introductions to the applications ChatGPT and the use of AI art, an overview of the terms of use and training data of popular AI image generators and the Mandelbrot debate in the context of the legal discussion. We shed light on the concerns and fears of press publishers in the article“Ancillary copyright and AI” and explore the fundamental question of whether AI can be an author in a separate article.