Describing LLMs

Objectives

After completing this lesson, you will be able to:

  • Describe LLMs
  • Identify benefits and risks of using LLMs

Introducing Large Language Models (LLMs)

Imagine that you want to use SAP AI Core to transform business data into intelligent, actionable insights across various sectors, including finance, HR, and supply chain management. Before diving into the world of generative artificial intelligence (AI), it is good to know some of the basic terms.

In this lesson, you will learn about Large Language Models (LLMs), and explore the benefits and risks of implementing LLMs in artificial intelligence systems.

An LLM is a type of artificial intelligence model that specializes in processing, understanding, and generating human language. These models are a subset of machine learning models known as deep learning models, which are designed to handle large-scale data and complex pattern recognition.

See the video to learn about LLMs, its capabilities, and benefits.

In-Depth Definition of LLMs

In this video, you will learn about a more detailed definition of LLMs along with their key aspects.

Benefits and Risks of LLMs

Once you have identified LLMs, you want to use generative AI to resolve your queries. Note that answers to your queries can be incorrect or may not present correct facts. It is advised tounderstand the benefits and risks associated with LLMs before you start using them.

In the context of SAP Business AI, LLMs have both benefits and risks that must be considered for businesses.

Benefits and Risks of LLM

BenefitsRisks
Efficiency: In the context of SAP Business AI, LLMs can significantly improve the efficiency of processes due to their ability to understand and process natural language at a large scale.Data Privacy Concerns: LLMs process a lot of textual data, potentially causing data privacy concerns. If sensitive data is processed, the system should ensure the confidentiality and anonymity of such data.
Cost Reduction: With LLMs, tasks such as customer support, data analysis, and others can be automated, thus reducing operational costs.Bias and Fairness: LLMs are trained on vast amounts of data from the internet so they might reflect and reproduce the biases present in those datasets.
Data Analysis: LLMs can analyze and interpret vast amounts of data faster and more effectively than humanly possible, providing businesses with valuable insights.Misinterpretation of Data: While LLMs can understand language, they can also make mistakes and misunderstand or misinterpret data, causing potential issues.
Improved Customer Experience: LLM-based applications can enhance customer experience by offering personalized assistance and real-time responses.Dependency: Excessive reliance on LLMs can make a business vulnerable if the system stops working or delivers inaccurate results.
Scalability: LLM can handle increasing amounts of work and interactions due to its deep learning capabilities.Technical Complexity: Implementing, fine-tuning, and maintaining LLMs requires technical expertise and resources. Mismanagement could lead to unexpected issues and costs.

Further Reading

  • AI Ethics at SAP(eLearning | 4 hours): ​Learn about the importance and principles of ethics in the context of artificial intelligence (AI). Discover how SAP implements AI ethics in the development, deployment, use, and sale of AI systems, and explore the concept of generative AI in a newly added course unit.
  • ​SAP Generative AI Cybersecurity Strategy​​(White Paper | 30 min) Learn about Generative AI (GenAI) and how it offers tremendous potential to improve the way the world conducts business, but also introduces new and complex security challenges, and how SAP is addressing both its use and its challenges.

Log in to track your progress & complete quizzes