Armin Fanzott in an interview with Confare.at
How important are data and artificial intelligence in companies today?
Artificial intelligence data and methods for using existing data already play an essential role in a wide variety of industries and areas of responsibility. For example, to predict customers with an affinity for switching (churn prediction), to derive maintenance measures (predictive maintenance) or for forecasts in sales (sales figures) or in purchasing (delivery quantities, inventories). Every company should address the question of how to convert existing data into use cases.
ChatGPT shows the potential that automation could go much further than we have experienced to date. AI receives intensive attention. Did you expect this development?
Yes, absolutely. ChatGPT receives so much attention because it is accessible to the general public and can be used by everyone. In this way, everyone can see the advantages and possibilities of the technology in a clear way. This was not the case with complex use cases such as predictive maintenance topics — these could often only be understood and implemented by experts.
Have the prospects for AI in companies changed as a result of the ChatGPT hype?
I don't think so. Even before ChatGPT, the topic was subject to a certain amount of hype and ChatGPT only reinforced it. Companies must continue to discover the potential of this technology and separate it from the hype. Even before ChatGPT, I had the feeling that companies were working intensively on the subject matter.
Will AI really be the job killer now?
Certain jobs will certainly fall victim to technology in the next few years. Perhaps also jobs that we considered safe a few years ago. Nonetheless, I think that jobs have also been created and will be created on the other side. With the rise of the car, no one needed a horse-drawn carriage anymore, so this job became obsolete. On the other hand, an entire automotive industry was created, in which Europe is now a pioneer. In the area of AI, there is still a lot of work ahead of us to catch up with the USA or China.
What new career paths are currently being created by data and AI?
Some, such as the data scientist, the data engineer or the machine learning engineer, have already emerged in recent years. However, I think that there will definitely be career paths of a less technical nature in this area in the next few years. For example, the UX designer with a focus on AI — who deals with user experience issues in applications that are based on AI.
For example: How do I teach a user of a skin cancer identification app that their birthmark looks very questionable? In other words, the AI has identified a high probability of carcinoma.
Which tools and options should you become familiar with now?
That depends heavily on the respective job position. There are new technologies in data science every month, so tool forecasting is extremely difficult. However, certain classics such as Python and associated libraries will certainly be with us for a while and you can't go wrong with that. In general, however, in addition to technological knowledge, a data scientist should also have business know-how.
In my opinion, it will be important in management to have a general overview of current AI/data science use cases in the respective industry and to know what skills I need in the company to implement them. Depending on your strategy, you can employ internal data scientists & data engineers together with software developers or use external support. Option 1 ensures a high level of knowledge development directly within the company, but sometimes leads to downtime or to data scientists being used more as BI experts due to a lack of use cases. In variant 2, you only employ them in the case of specific projects and can concentrate more on your core business. A combination of mixed teams of internal experts and external support in the event of resource bottlenecks is definitely also an opportunity to make efficient progress.
What role should AI now play in a company's digital strategy?
An essential one. In my opinion, no digital strategy can do without AI. I even go one step further and say that there should be no digital application without at least considering data science or AI use cases. We take a closer look at this subject area in every project concept so that potential can be considered at an early stage.
What contribution can corporate IT make in this regard? What is the distribution of roles within the company?
The distribution of roles depends on the type of organization. Some companies create central AI/data science departments that should uniformly process all use cases across the company. Others rely on distributed teams in the respective departments and guilds or communities to exchange ideas with each other. Option 1 offers major advantages in terms of uniform technology orientation and infrastructure, but on the other hand, in my opinion, creativity suffers as use cases are mostly driven by the specialist sector and may then never reach the central unit. Option 2, on the other hand, provides direct contacts in the specialist areas who plan and implement cases without bureaucracy. On the other hand, I can say from experience that technical exchange among data scientists can suffer greatly from this variant.
I generally advise small to medium-sized companies to use external implementation partners rather than to set up their own organizational units, as the large number of use cases usually does not exist to justify such a structure. Individual projects can be implemented here more cost-effectively with external support, so that you can still benefit from AI solutions.
In both cases, corporate IT must provide a flexible, modern and secure infrastructure and tool landscape and act as a technical contact for data scientists and data engineers.
.jpeg)
