What type of legal protection could be obtained for a novel software algorithm?
Algorithms play an increasing role in creative production and consumption. An algorithm is nothing more than a method for solving a specific problem or completing a task, which can include a basic set of rules/instructions as well as machine learning and more cutting-edge artificial intelligence (AI) code. Show
There are many examples of the production of creative goods and services involving the use of algorithms. AI applications that have been developed are capable of creating artistic, musical, and literary works (such as paintings, music, poems and news articles) with very limited or even no human intervention in the final product. Researchers increasingly rely on data mining algorithms to handle a large amount of data, which may include copyright works, and to be able to extract patterns, trends and associations and gain insightful knowledge. Creators of video games also employ a broad set of algorithms to respond to players’ inputs. They use algorithms not only to mimic simple human behaviours, for example, the ghosts’ movements in the Pac-Man game, but also to formulate more complex and sophisticated behaviours, such as the life simulation of the characters in The Sims. Moreover, music streaming services such as Spotify and Amazon Music employ algorithmic systems to recommend artists, bands, and new songs to their listeners. In a similar way, user-generated content platforms, such as YouTube and TikTok, use algorithmic systems to recommend as well as to moderate the content that circulates on their platforms (as further explained in the Copyright User UGC page). Even though algorithms are not something new, given their increasing significance in the creative industries, it is important that users have a better understanding of how to deal with their influence on the production, distribution and consumption of creative works and the implications for copyright. Algorithms and CopyrightCopyright law in the UK, as elsewhere, does not protect ideas, methods, or procedures. However, the expression of an idea can be protected if it is original and falls into one of the categories of protected works. Therefore, in order to be eligible for copyright protection, programmers need to convert the algorithms into a source code, which, as a fundamental component of a computer programme (software), can be protected as a literary work by copyright law. Copyright protection prevents others from copying the code without the copyright owner’s permission but not from writing a different code with the same function. One of the most common categories of algorithms used in computer programming is collaborative filtering algorithms, which are normally employed in the recommendation systems used by music streaming services. They provide personalised music experiences to their users based on the opinions of other users who share similar characteristics and music preferences. These algorithms analyse past user behaviour in order to establish connections between users and songs and to predict which songs or playlists each user would like. In some jurisdictions, programmers can also protect algorithms as patents. In this case, they need to connect algorithms and their frameworks with some industrial or technical application. This means that, although mathematical algorithms per se are not patentable, inventions encompassing mathematical methods, along with computer programs, are patentable as long as they present a technical contribution. However, some producers might prefer to keep their algorithms secret, protecting them as trade secrets. As an intellectual property asset, algorithms are usually proprietary, which means programmers or the companies they work for reserve the rights to use, modify or share these algorithms. However, as algorithms are rarely deployed alone, they are often incorporated in AI tools or software. These AI tools can also be licensed as open source software. In these cases, not only the algorithms but also their core statistical models are made open source. When developing software, for example, programmers usually incorporate the algorithms (the set of instructions) as actual code. If the software is released as open source software, where the underlying code is publicly available for anyone to access, use or modify without restrictions, the embedded algorithms are also available to be used freely by any programmer, data scientist, or anyone else who wish to make use of them.
|