The world of artificial intelligence (AI) has evolved rapidly with the introduction of generative models such as ChatGPT. These innovative approaches to AI are not only revolutionising the way we interact with information, but are also transforming a wide range of industries by automating complex tasks. In this blog post, we explore the fascinating use cases of generative AI (GenAI) and look at the future possibilities of this technology.

4.4 trillion dollars: GenAI will change the world of work

In its study "The economic potential of Generative AI" by 2023, McKinsey & Company describes an increase in the value of the global economy of $2.6 to $4.4 trillion - figures that can only be achieved through extensive application potential across all industries and processes. According to the study, up to 75% of this potential lies in the areas of customer operations, research and development, software engineering, and sales and marketing. In particular, the language or syntax intensive and creative tasks that are often found in these areas can be supported and partially automated through the use of language models.

Here we take a look at five of the most common use cases for generative AI.

Intelligent document processing: more than just OCR

Intelligent document processing (IDP) is a core area where generative AI offers immense benefits. Modern IDP systems go far beyond traditional optical character recognition (OCR). They recognise not only text, but also the structure and context of documents. Solutions such as Tamed AI recognise documents in three dimensions and extract information accurately, significantly increasing efficiency when dealing with correspondence, faxes and sometimes handwritten notes.

One example of the use of AI is in the automation of invoice processing. Generative AI can be used to capture, validate and process invoices in different formats, languages and currencies. This reduces manual effort, speeds up payment processing and improves accuracy. In addition, AI can extract and analyse information from invoices to identify patterns, detect anomalies and provide valuable business insights. This helps organisations reduce costs, increase liquidity and strengthen compliance.

Domain Knowledge Agents: Specialists in specialised knowledge

Domain Knowledge Agents are specialised applications of generative AI that are trained to act not only as a source of information, but also as active participants in specific domains. These agents can assimilate extensive knowledge in a specific domain and use it in a way that goes beyond the mere reproduction of information. They can use this knowledge to generate new insights, answer complex questions, and even formulate and evaluate hypotheses.

A key advantage of domain knowledge agents is their ability to provide contextual and in-depth answers that are at the forefront of the field. This makes them a valuable resource in fields such as medicine, law, engineering and other scientific disciplines where continuing education and access to up-to-date information are critical.

A practical example of the use of domain knowledge agents is in medical diagnostics, where they can help doctors analyse symptoms and identify possible treatment approaches. In legal practice, they can assist lawyers in researching case law and legal texts, thereby increasing efficiency. In academic research, they serve as assistants, conducting extensive literature searches and helping researchers develop new studies.

Software engineering: from analysis to deployment

Application development also benefits from the capabilities of generative AI. AI models can support everything from requirements analysis and code development to the operation of software. They can help transform legacy code, support system migration and provide insight into existing software architectures. This enables development teams to organise their work more effectively and efficiently.

One example of the use of GenAI in application development is the automatic generation of documentation. By analysing existing code and comments, AI models can generate descriptive text that explains the functionality and design of the software. This helps both the developers themselves and other stakeholders who need to interact with the software. Documentation can also be tailored for different audiences, such as technical or non-technical users. Automated documentation saves time and resources that would otherwise be spent on this tedious and often neglected task.

Creative writing: Hyper-personalisation and style adaptation

In creative writing, generative AI makes it possible to adapt language and style, shorten or lengthen text, and perform sentiment analysis. These capabilities are particularly useful in marketing, where hyper-personalisation and targeted content are becoming increasingly important.

A concrete example of the use of generative AI in sales or marketing is the creation of personalised email copy. Generative AI models can use existing customer data, such as purchase history, interests or demographics, to generate tailored messages that increase customer loyalty and conversion rates. Texts can also be tailored to achieve different goals, such as attracting new customers, encouraging repeat purchases or promoting special offers.

Structured data and data storytelling

Generative AI models can also process structured data and translate it into natural language. They can build SQL queries and not only retrieve data, but also interpret it. Data storytelling becomes a powerful tool for presenting data analysis in an understandable and accessible way.

An example of how generative AI can be used in data storytelling is the creation of automated reports on the energy consumption and emissions of different customers or regions. Generative AI models can analyse data from smart meters, sensors or other sources, identify trends, extract key insights and convert them into understandable and accurate text. These texts can then be used as the basis for decisions, presentations or further analysis.

And so much more

So far we have focused on text-to-text capabilities, but the list of use cases is far from exhaustive. However, when we look at the broad field of multimodal language models, new possibilities for creative design, communication and learning open up. Generative AI models can transform not only text to text, but also text to image, text to audio and text to anything else. More on this in another blog post.

Conclusion and outlook

The use cases for generative AI are diverse and can offer different benefits depending on the application and objective. It is important that organisations experiment with these technologies and identify the best use cases for their specific needs. Integrating generative AI into business processes promises to optimise workflows and increase efficiency. However, it is important not to neglect the human element and to use the technology responsibly. As we continue to push the boundaries of what is possible with AI, we should ensure that we use the technology for the benefit of all, while empowering the human factor.

Picture Tim Bunkus

Author Tim Bunkus

Tim Bunkus has been an expert in artificial intelligence (AI) and machine learning since 2015. He advises clients on the holistic integration of AI into the individual business context. With broad methodological and technical knowledge and extensive project experience in various roles (consulting, project management, architecture), he is able to design use cases and implementation strategies and evaluate them.

Save this page. Remove this page.