AI-generated picture

For an innovative company, it is crucial to remain curious and open to new developments. Therefore, the trending topic of Artificial Intelligence, or "AI" for short, has naturally not passed by Polytives unnoticed. We have read reports, examined use cases, and even witnessed firsthand how the impact of AI must be considered when implemented in the production process. We have also participated in various webinars and events on the subject, such as those offered by the Chemistry Cluster Bavaria or the TITK in our neighborhood. Below, we would like to provide an overview of what we have learned so far, with a focus on AI in research and development.

Experiment design, data analysis, documentation

An unbeatable advantage of AI is the simplified experiment design, which supports statistical experimental design methods. This can save a great deal of time when analysing different variables and parameters. Experiments can also be automated and carried out by AI-controlled robots.

The amount of data generated in this way can then be analysed - here, AI recognises patterns, trends and complex relationships and dependencies, allowing new synthesis routes to be identified and implemented more quickly and easily. When it comes to molecular modelling and simulation in particular, AI is making major contributions to research and development. The aim is to change molecular structures in such a way that desired properties are specifically adapted - an important means of developing medicines or, for example in the industrial sector, investigating material properties such as strength or thermal stability.

AI in research and development can also touch on an area that is rarely considered outside of academia: The publication of results. The publication of scientific articles is an elementary component of the accumulation and dissemination of knowledge. Here, AI can help not only with literature research, but also with the publication of articles in general. Even if people are sceptical about direct publishing, there are already promising reports that AI-supported text structuring facilitates the scientific analysis and peer review of manuscripts.

Transfer to industry

In order to transfer the fields of application of AI in research and development to industrial issues, partners are currently needed who generate a large amount of data and test and evaluate the processes derived from it. Irrespective of this, applications in the areas of quality assurance, process control, etc. have of course already emerged. But AI can be a possible basis for linking academic research fields and industrial expertise - this has not yet been done to this extent. However, we are sure that innovative ideas will always seek ways and means to reach the world and would not be surprised if there is significant progress in this area in the coming years.

We will definitely stay on top of it and test, wherever possible and practical, how the positive effects of AI tools can support our daily and not-so-daily challenges. An example outside of AI in research and development is the use of AI in marketing. For instance, ChatGPT is quite useful for illustrating a blog post.