How industrial companies create an AI culture
The use of artificial intelligence in a company only remains valuable if all teams involved have access to the data. Cross-cutting collaboration with one tool and a future-proof AI culture is the key.
“At a certain point, access to data and artificial intelligence (AI) projects needs to be opened up to employees outside of data teams.” So says Rachel Boskovitch, an expert at the enterprise AI platform Dataiku. If they don’t, industrial companies will reach the limits of the potential of this future technology. Only when, in addition to highly qualified data scientists, decision-makers, business profiles, and employees from a wide range of company departments are involved, will better developments emerge. This requires a central tool that enables real-time data access and a new level of cross-team collaboration.
New opportunities with holistic AI culture
For employees from different departments to be involved in working with data, a holistic AI culture is necessary within the company. Only if all teams can identify with working with data and have a fundamental understanding of the resulting added value will the new type of collaboration work.
The holistic AI culture is strengthened by tailored training that is necessary to work with data. Only in this way can the different teams understand to what extent and for what purpose AI is implemented in projects and what the expectations and challenges are. Emerging misunderstandings can be prevented.
In this transformation, not all employees need to have a deep understanding of big data, data science, or artificial intelligence. It is sufficient if the employees involved know where data comes from, what it is used for, and how the data workflow is structured. Once this fundamental step has been taken, further opportunities arise beyond existing use cases: The perspectives gained and the associated creativity of the individual positions are the basis for new approaches to data use, which are further developed into real applications by AI experts.
Citizen Data Scientist
When new teams are integrated, new positions are created in the company at the same time. The best example of this is the Citizen Data Scientist. Unlike the highly qualified Data Scientist, the Citizen Data Scientist is primarily familiar with the problems and requirements in the industry and knows the business requirements for data projects in detail. Although this position does not require in-depth knowledge of machine learning, it is possible to be actively involved in working with data and creating models – without having to rely on a Data Scientist.
Integrating knowledge from non-tech departments into projects
With the democratization of AI in the industry, a new, increased collaboration and communication across all company divisions and positions automatically result. This means that the knowledge from individual specialist departments can be actively integrated into projects, which contributes to better solutions through new perspectives. In the future, important insights into processes will no longer be missing, nor will there be any information gaps.
Existing resources will be used much better. Highly qualified Data Scientists are relieved by Citizen Data Scientists and the use of AutoML. Instead of basic work such as data preparation, Data Scientists can focus on highly technical and specific issues and their solutions.
For industrial companies, the democratization of AI has very practical advantages: Less time wasted, better and new AI developments, and last but not least, the transformation into a future-proof, data-driven company. Although this requires investments and a lasting change in corporate culture, the industry is at the same time getting ready for a more value-added, data-driven future.
Author: Roland Reiter