SAP Integration Suite is a cloud-based integration platform that enables businesses to seamlessly connect and integrate applications, data, and processes across diverse systems, both on-premise and in the cloud.
It has introduced a groundbreaking new feature: Generative AI Based Integration Flow Generation, powered by SAP AI.
This functionality harnesses the power of SAP Artificial Intelligence to streamline and accelerate the creation of integration flows, allowing users to simplify complex workflows with ease.
By automating the process of building these iflows with SAP AI we can reduce the need for manual intervention, automate tedious tasks, reduce errors, and increase productivity of consultants.
This marks a significant leap forward in how SAP Integration Suite supports AI-driven innovations for seamless and scalable integrations.
What is AI-Assisted Generation of Integrations?
SAP Integration Suite is offering a powerful AI feature in the suite to create integration flows(iflows). Once the feature is activated, SAP integration consultants can leverage AI to generate integration flows.
The consultant simply provides a description of the integration scenario when prompted, and the Generative AI tool uses that input to automatically create the corresponding integration flow.
The AI-driven approach is simplifying the development process, enabling faster and more efficient creation of integrations based on specific requirements.
Let’s dive in and create our first integration flow using Artificial Intelligence.
Enable Generative AI Feature
Step 1: Logon to the integration tenant.
Step 2: Go to Settings>Integrations>Generative AI. And choose Edit.
Enable the Activate Generative AI Features checkbox.
Next, click Save.
Generate Integration Flows with AI
Step 1: You need to add relevant systems like S/4Hana under System Landscape in SAP BTP Cockpit.
Once a successful SAP Integration Suite formation is established with relevant systems, the Generative AI-Based Integration Flow Generation feature automatically scans the systems and their APIs.
The systems and APIs will be visible in the Sender and Receiver sections of the Generative AI-based integration flow creation UI once the above is configured. This capability makes it much easier to select and link systems, simplifying the entire process of building iflows.
For detailed guidance on enabling and configuring System Landscape in SAP BTP Cockpit for SAP Integration Suite, please refer to the documentation.
Step 2: From the Integration tenant Navigate to Integrations and APIs > Edit.
Step 3: Click on Add > Integration Flow.
Step 4: Choose Generate Integrations with assistance from AI.
Step 5: Describe the integration scenario and Click Send. You can also choose a sender and receiver system.
Step 6: Click Generate .
Congrats. you’ve now generated an iflow using the Generative AI-Based Feature in SAP Integration Suite.
Summary
In this blog, we’ve explored the powerful new Generative AI-based integration flow generation feature in SAP Integration Suite, which leverages SAP AI to automate and streamline the creation of integration flows.
By reducing manual effort and simplifying complex workflows, this feature enhances productivity for consultants and ensures more accurate, efficient integrations.
The step-by-step tutorial provided demonstrates how easy it is to activate and use this functionality, showcasing how AI-driven innovations are shaping the future of SAP Integration Suite for seamless, scalable business operations.
In this blog, we’ll explore SAP Joule and understand how it helps businesses streamline processes, improve decision-making, and increase productivity of the business users.
SAP Joule is more than just an AI tool—it’s a game changer. It can provide real-time, contextual insights directly within SAP applications. Let’s dive deeper into its functionality, architecture, and its potential for business transformation.
What is SAP Joule?
SAP Joule is a generative AI assistant embedded across SAP’s enterprise software ecosystem. It acts as SAP’s AI copilot, allowing users to interact with SAP systems through natural language—just like you would chat with a colleague or friend.
You can ask SAP Joule to retrieve information, execute transactions, or provide insights, all within the context of your SAP environment, whether it’s SAP SuccessFactors or SAP S/4HANA Cloud.
SAP Joule is provided out of the box and natively integrated in SAP’s cloud applications. It comes with pre-configured content that is active right away for customers with limited configuration effort.
Currently, Joule is embedded in SAP cloud applications like SAP SuccessFactors and SAP S/4HANA Public and Private Cloud, and more applications to follow. However, keep in mind that Joule is available only for cloud solutions, not on-premises customers.
Why SAP Joule is a Game-Changer?
Remember the days when SAP users had to memorize complex Tcodes? Later, Fiori apps simplified things. Now, SAP Joule takes it to the next level.
So, what makes SAP Joule stand out?
Context awareness: Joule provides intelligent responses based on the user specific SAP transactions, and customer business data. It can maintain the context across different SAP products, whether you’re accessing Joule through SAP Start or directly within an application like SAP Successfactors.
Grounded on business documents: SAP Joule uses techniques like Retrieval Augmented Generation (RAG) to link to relevant sources, such as internal policies or business records, to provide accurate information and prevent AI from making things up.
Security and compliance: SAP Joule follows user authorization and authentication rules, keeping sensitive data safe and ensuring everything stays compliant.
Guardrails for responsible AI: It’s designed with safeguards against inappropriate use, including bias or hate speech.
Seamless Integration: Joule is built into the existing SAP environment, meaning it requires minimal setup and training. It integrates naturally with the systems you’re already using, making it highly accessible.
How Does SAP Joule Work?
Lets understand the high-level architecture of SAP Joule.
The user starts with their query in the Joule client in respective SAP application.
Each incoming request is processed based on 3 categories:
Scenario Catalog: Joule analyzes if the user-prompt is relevant for Joule’s Scenario Catalog. This catalog contains metadata of all available scenarios, functions, and skills of SAP cloud applications.
Knowledge Catalog: Then it conducts an informational filtering based on the Knowledge Catalog. This contains SAP-knowledge as well as the customer-owned knowledge. This process is based on Retrieval Augmented Generation (RAG).
Context: SAP Joule is aware of the user’s context and history. This information includes which SAP cloud application the user is using and which additional SAP license the customer has. Also, Joule is aware of the user’s role(s) and permissions. This means a user cannot access information or execute business processes which they are not authorized to do. Lastly, SAP Joule understands a user’s chat history and context.
Next, SAP Joule takes all of this information and gives a much richer query to the LLM which is responsible for the dialog management.
The LLM provides a grounded response back to the Joule service.
In this case, Joule calls the respective SAP backend system to proceed with the requested query, for example requesting for leave in successfactors.
The response is then filtered, and the customer gets a reply with full details that the conversation and output have been handled securely, with enterprise-level security and data privacy in place.
Great, now we understand that SAP Joule is an AI Copilot that’s integrated with SAP solutions to chat with it to retrieve information or perform tasks.
Use Case: AI Assisted Person Insights
An easy to understand use case of SAP Joule is the AI feature in SAP’s HR solution SAP SuccessFactors – AI-Assisted Person Insights.
As a manager, salary discussions can be one of the most challenging tasks. And, SAP SuccessFactors helps managers navigate these tough conversations with AI-assisted insights, ensuring they are well-prepared and confident.
For instance, SAP Joule, the AI co-pilot, helps managers review detailed insights on employees. It provides details about the employee’s compensation history, promotion record, bonus eligibility, and career development. The AI can even compare the compensation with market rates, making salary discussions more data-driven and balanced.
SAP Joule is much more than that, it can even help software developers write code and accelerate productivity using Generative AI. Let’s see a demo!
Demo
Let’s dive into a practical example of SAP Joule’s Generative AI and show you how to use it with SAP Build to develop a full-stack application.
By the end of this demo, you’ll know how to:
Create a full-stack project using a template,
Develop data entities and services with Joule’s Generative AI,
Generate application logic using Joule’s AI capabilities, and
Add a UI to the application and perform testing.
Watch the demo here:
So, What’s Next? Joule Extensibility
Many customers are asking to extend Joule’s capabilities. SAP is enabling users to create their own Joule AI skills via SAP Build, offering a low-code approach to developing custom skills.
Here are some exciting developments coming soon:
Joule Studio in SAP Build: Joule Studio is planned for Q1 2025. allowing users to create and manage their own skills. It is a new tool offering drag-and-drop simplicity for building, deploying, and managing custom skills.
Third-Party Integration: Joule will be able to integrate third-party systems, offering a fully integrated conversational experience.
Autonomous AI Agents: Soon, you’ll be able to build custom autonomous AI agents to solve complex workflows and collaborate with other agents in Joule.
Summary
SAP Joule is transforming the way businesses operate, offering AI-driven solutions that go beyond simple automation. SAP Joule is at the heart of SAP’s AI strategy, and as we move forward, its role will only become more critical in shaping the future of enterprise operations. Whether you’re in HR, finance, supply chain, or customer experience, Joule has something to offer that can radically change how you work.
SAP AI uses various AI methods, including generative AI models to empower organizations to automate processes, enhance decision-making, and generate insights from massive datasets.
Generative AI plays a significant role in SAP AI’s capacity to enhance customer experiences, and develop intelligent assistants like SAP Joule.
In this blog, we will dive into the core components of Generative AI and how these models operate – allows us to not only discuss the latest advancements but also implement them effectively in real-world SAP environments.
How Does Generative AI Work?
Generative AI allows machines to generate new, and original content. It could be text/images/videos. In the enterprise world there are many applications to it. A simple example is if you’re using the project module of SAP, A Gen AI feature within the software can generate a project summary based on the current status of the project by analysing the budget/timeline/expense/tasks etc from the system. Otherwise as a PM you’re required to create an executive summary yourself by looking at multiple screens and data. Its immensely improving the productivity.
Generative AI uses deep learning techniques like neural networks to understand patterns in large datasets. Lets understand what are the key methods Gen AI uses to create content:
1. Generative Adversarial Networks (GANs)
One of the most revolutionary approaches in generative AI is the Generative Adversarial Network (GAN). GANs consist of two neural networks working in opposition:
Generator: This network generates new content, like an image of a face. The goal of the generator is to create data that is indistinguishable from real data.
Discriminator: The role of the discriminator is to evaluate whether the generated content is “fake” (generated) or “real” (from real datasets). Over time, the discriminator improves at distinguishing between real and fake, forcing the generator to create more realistic data.
These networks are locked in a constant competition, resulting in highly realistic outputs over time. For example, in image generation, the generator creates images that the discriminator judges as fake or real until the generator becomes proficient at mimicking realistic images.
This adversarial process has been groundbreaking, especially in fields like computer vision, where GANs can generate photorealistic images, transform images into different styles, or even create entirely new ones.
2. Variational Autoencoders (VAEs)
Variational Autoencoders (VAEs) are another significant deep learning model in the generative AI landscape. Unlike GANs, VAEs focus on learning a compressed representation (or latent space) of the data. The process involves:
Encoder: Compresses the input data (e.g., an image) into a smaller, encoded representation.
Latent Space: A compressed version of the input that holds the essential features of the data.
Decoder: Reconstructs the original data from the compressed representation.
What makes VAEs unique is their ability to generate new data based on the learned latent space. For instance, after learning the core features of human faces, a VAE can generate entirely new faces that share characteristics of the faces in the dataset.
VAEs are useful in generating diverse outputs, ranging from new visual content to potential designs, and are known for their ability to interpolate between examples smoothly.
Conclusion
Generative AI is transforming industries by enabling machines to generate realistic, high-quality content. Techniques like GANs and VAEs are pushing the boundaries of what AI can do, making it an essential tool for enterprises.
Let’s explore together how SAP AI and generative models can drive your business forward. Keep learning!
The rapid advancements in artificial intelligence (AI), particularly with large language models (LLMs), have opened doors to more intelligent business solutions.
However, challenges like hallucination—where models provide incorrect or misleading information—present significant hurdles.
A key innovation addressing this issue is Retrieval-Augmented Generation (RAG). RAG combines the strengths of LLMs with real-time data retrieval, offering businesses a practical and cost-effective solution for contextual, grounded insights.
In this post, we’ll explore what RAG is, why it’s needed, and how SAP is enabling its customers to harness its power with real-life examples.
What is Retrieval-Augmented Generation (RAG)?
RAG is a framework that enhances traditional generative AI by integrating a retrieval mechanism.
When an AI model receives a query, it retrieves relevant information from a database or document repository, allowing the model to generate responses grounded in real-time, up-to-date information.
This process drastically reduces the risk of hallucinations, making the AI outputs more relevant and trustworthy for business applications.
In simple terms, RAG brings together two main components:
Retrieval: Accessing relevant documents or data in response to a query.
Generation: Using that retrieved information to generate a coherent and contextually appropriate answer.
Why is RAG Needed?
While LLMs have revolutionized AI-driven insights, they still face several limitations in real-world applications, particularly for businesses:
Hallucinations: LLMs may provide factually incorrect or irrelevant answers, especially when they’re disconnected from real-time or domain-specific data.
Costly Fine-Tuning: Training or fine-tuning LLMs with business-specific data is complex, expensive, and time-consuming.
Static Knowledge: LLMs, once trained, do not automatically update their knowledge, making them less useful for industries that rely on dynamic data.
By embedding a retrieval process, RAG reduces these challenges, ensuring that AI-generated responses are not only relevant but also grounded in the most recent and accurate data.
Why is RAG Relevant for SAP Customers?
SAP’s customers span industries such as supply chain management, HR, CRM etc, all of which depend heavily on accurate, up-to-date information.
For these businesses, the integration of RAG can significantly improve decision-making by:
Delivering reliable, grounded insights: With RAG, LLM outputs are based on the latest business data, offering contextualized and reliable responses.
Reducing response times: Business workflows often involve searching for information across various documents. RAG enables conversational AI tools to fetch this information in seconds.
Lowering Costs and Complexity: Instead of expensive fine-tuning of LLMs, businesses can use their existing data repositories to power RAG models.
How SAP is Enabling RAG
SAP has taken significant steps to incorporate RAG into its ecosystem, allowing businesses to build and deploy AI solutions grounded in their own data. Here are key elements of SAP’s approach:
SAP’s Central RAG Service
SAP provides a central, domain-agnostic RAG service that integrates with user interfaces, offering:
A scalable method to ensure AI systems use the most up-to-date information.
A framework to trace information sources, which helps verify AI-generated answers and build trust with end-users.
A vector engine in SAP HANA Cloud to store and query data embeddings efficiently .
Document Grounding in Joule and Generative AI Hub
SAP’s AI services like Joule—an AI copilot—incorporate document grounding, a core RAG technique:
Indexing: Documents are ingested, split, and stored as vector embeddings.
Retrieval: The system retrieves relevant data chunks for user queries.
Generation: This information is used by LLMs to create relevant and accurate responses .
Business Use Case: Using RAG in SAP SuccessFactors
Consider an HR team using SAP SuccessFactors. Traditionally, responding to policy-related inquiries could take 20 minutes per question, creating a heavy workload.
By implementing RAG through Joule, employees can now self-serve by querying policy documents directly. The system retrieves the most relevant document sections and generates a concise answer, reducing HR inquiry volume by 35% .
Conclusion
Retrieval-Augmented Generation (RAG) is revolutionizing the way businesses leverage AI by ensuring that AI-generated responses are contextual, relevant, and reliable.
SAP’s integration of RAG across its platforms, such as Joule and BTP, makes this technology accessible to its wide range of customers.
By reducing the need for costly LLM model fine-tuning and minimizing hallucinations, SAP is paving the way for its customers to benefit from powerful, trustworthy AI insights in their daily operations.
At SAP TechEd 2024, SAP made several major announcements regarding its advancements in Generative AI. With a focus on simplifying business processes, enhancing customization capabilities, and bringing AI-driven insights to the forefront, SAP is paving the way for intelligent enterprise solutions powered by AI. Here’s a roundup of the key announcements that will reshape how businesses interact with SAP’s ecosystem.
1. SAP Build’s Generative AI Toolkit for SAP HANA Cloud
SAP has introduced a powerful Generative AI toolkit for SAP HANA Cloud, embedded within SAP Build. This new toolkit allows users to perform common data analysis tasks and handle even the most complex machine learning operations—no deep technical expertise required.
With the AI Toolkit for HANA Cloud, you can now leverage natural language prompts to execute sophisticated tasks. For example, if you need to create a time-series forecast model, you can simply ask Joule—SAP’s AI assistant—using plain language, and it will handle the rest. This reduces the need for understanding the intricacies of machine learning, making it easier for business users to incorporate AI-driven insights into their processes.
The integration of Joule and SAP Build Code simplifies the user experience, ensuring that even those with little to no coding experience can harness the power of AI in SAP HANA Cloud.
2. AI-Generated Recommendations in SAP Integration Suite
Another exciting update is the introduction of AI-generated recommendations in the SAP Integration Suite. When working on custom scripts, users can now open the script editor and generate optimization proposals with the help of Generative AI.
This new capability makes it possible to enhance and optimize custom integration scripts, ensuring that processes run more efficiently. By leveraging AI, users can save time and avoid common pitfalls, allowing them to focus on higher-level business logic instead of getting bogged down with technical details.
3. Joule Studio: Creating Custom Skills with Low-Code AI
Joule Studio, now part of SAP Build, is designed to unlock the power of Joule in building custom skills that integrate seamlessly with your business landscape. Joule Studio is a low-code environment that enables developers and business users to design and deploy AI-driven skills without needing to write complex code.
For example, if you want to pull service tickets from ServiceNow into Joule, you can use Joule Studio to create a custom skill with ease. By configuring the necessary parameters in ServiceNow APIs, you can deploy the skill and instantly access the data from within Joule. Similarly, you can integrate SAP SuccessFactors data and other business-critical information.
This capability empowers business users to create tailored applications that enhance decision-making and streamline operations. Joule Studio will be available in beta later this year and will become generally available (GA) soon after.
4. Generative AI Hub: New Models and Enhanced Capabilities
SAP’s Generative AI Hub is receiving significant upgrades, offering a broad array of state-of-the-art closed and open-source models, out of the box, integrated with SAP’s applications. This will give developers access to the best of both worlds: SAP’s real-time business data combined with foundational AI models from Mistral and Meta.
The Generative AI Hub will also introduce: – Orchestration services for managing multiple AI workflows. – Data masking and content filtering to ensure privacy and security. – Prompt templates and prompt management capabilities, making it easier to deploy AI-driven applications across different use cases.
Furthermore, the hub will come with SDKs for Java, JavaScript, and ABAP, ensuring that developers across different technologies can leverage SAP’s AI tools effectively.
5. SAP Knowledge Graph with SAP HANA Cloud Knowledge Graph Engine
The SAP Knowledge Graph is one of the most transformative tools in SAP’s AI offering. This new engine enables developers to ground Large Language Models (LLMs) in the entire semantic model of SAP, unlocking new potential for building intelligent business applications.
The SAP Knowledge Graph encompasses: – Over 450,000 ABAP tables. – More than 80,000 CDS views. – Over 7 million properties and fields.
This vast repository of structured metadata is essentially the bridge between natural language processing (NLP) and SAP’s highly structured data model. It allows developers to create more flexible use cases, such as advanced information retrieval, AI-based predictions, and analytics—all grounded in SAP’s data model.
By leveraging the SAP HANA Cloud Knowledge Graph Engine, developers can build business applications like never before, with real-time access to SAP’s rich data environment.
The SAP Knowledge Graph and its engine will be generally available in Q1 2025, offering unprecedented capabilities for AI-driven innovation within SAP landscapes.
Looking Ahead
SAP’s announcements at TechEd 2024 signal a major leap forward in how Generative AI will drive digital transformation. By combining cutting-edge AI models with SAP’s real-time business data, these new capabilities make it easier than ever for businesses to create smarter, more efficient processes.
From enabling non-technical users to generate machine learning models in SAP HANA Cloud to providing developers with advanced tools to create custom skills and integrate AI-driven insights, SAP is truly unlocking the power of AI across its ecosystem.
Stay tuned for more updates as these capabilities become available, and get ready to take your SAP solutions to the next level with Generative AI.
In our previous blogs, we explored various fundamental concepts surrounding artificial intelligence, including an introduction to AI, machine learning, deep learning, generative AI, large language models (LLMs), and prompt engineering. Each topic laid the groundwork for understanding the advanced capabilities of AI and its applications across industries.
Now let’s understand what’s SAP’s AI strategy.
This 7 part blog series is designed for anyone who wants to start learning about SAP AI. No matter what your job is or what you do, understanding the basics of Generative AI how it connects to SAP is a smart choice.
In this blog series, we’ll break down SAP AI into easy-to-understand parts, covering one step at a time.
SAP’s approach to AI is structured around its AI Strategy, which aims to infuse AI into applications and business processes through two main pillars:
Embedded AI in SAP Applications
Exposing Business-Centric AI Services via SAP BTP
Embedded AI in SAP Applications
SAP integrates AI capabilities natively within its business applications such as SAP S/4HANA, SuccessFactors, SAP Fieldglass etc. This allows customers to leverage AI without needing extensive development efforts.
For example, AI-assisted person insights in SuccessFactors helps managers identify discrepancies in employee compensation by analyzing historical data.
Similarly, the SAP Integration Suite now automates the creation of integration flows based on user-defined scenarios, streamlining the integration process.
In SAP Fieldglass, generative AI simplifies recruitment for external workers, enabling hiring managers to generate job descriptions 85% faster and project managers to deliver statements of work more efficiently.
Exposing Business-Centric AI Services
SAP is focused on offering targeted AI services through SAP Business Technology Platform (BTP) for custom scenarios. By partnering with AI leaders like OpenAI, NVIDIA, Microsoft etc SAP provides access to foundational models, enabling developers to build custom applications featuring generative AI.
Notably, SAP leverages Microsoft’s Azure OpenAI Service, allowing customers access to powerful LLMs without needing individual contracts with third-party vendors.
What is SAP Business AI?
You may have heard about SAP Business AI in multiple occasions and we want to simplify it for you. SAP Business AI encompasses a suite of AI-related solutions aimed at unifying SAP’s offerings.
You can image it as an umbrella under which all AI related solutions are included. It’s just an idea but not a single product. It includes multiple components.
Key components are:
Joule: An advanced generative AI copilot embedded throughout SAP’s cloud enterprise portfolio, delivering contextualized insights and enhancing productivity in a secure manner.
Embedded AI Capabilities: As discussed earlier these capabilities are embedded across SAP solutions, offering functionalities like intelligent document processing, personalized recommendations, and forecasting.
AI Foundation on SAP BTP: A collection of ready-to-use AI services and tools designed to accelerate AI integration into business processes.
What About SAP’s Own Foundation Models?
We’ve seen how SAP is using other foundation models like OpenAI’s to build AI solutions. But is SAP coming up with their own LLMs? the answer is Yes!
SAP is developing the SAP Foundation Model, a table-native AI model designed for predictive tasks on tabular data. Unlike LLMs, which excel in text generation, the SAP Foundation Model aims to outperform traditional methods in classification and regression tasks, thus ensuring reliable and relevant AI solutions tailored to business needs.
Conclusion
SAP Generative AI is not just about leveraging advanced technologies; it’s about embedding AI seamlessly into business processes and applications. By providing users with powerful tools and insights, SAP is committed to helping organizations unlock the full potential of AI to drive better outcomes.
Stay tuned for more insights as we continue our exploration of SAP Generative AI and its transformative impact on the business landscape.
In this blog, we’ll guide you through the process of building a full-stack application from scratch using SAP Build Code and SAP Joule’s Generative AI.
We’ll leverage SAP Joule’s Generative AI to streamline the development of data models and application logic for essential functionalities like customers, purchases, and redemptions. Whether you’re a seasoned developer or just starting out, this walkthrough will show you how these powerful tools can accelerate your development process and enhance your projects. Let’s dive in!
Before diving in, here’s what we’ll cover in this guide:
Create a full-stack project using a template.
Develop data entities and services with Joule’s Generative AI.
Generate application logic code using Joule’s Generative AI capabilities.
Add a UI to the application and perform testing.
Step 1: Logon to SAP BTP Trial. If you don’t an account and want to setup one this blog will help.
Once your trial account is ready follow these steps: Click on “Go To Your Trial Account”.
Step 2: Click on Boosters.
Step 3: Search for “Get started with SAP Build Code ” and Click on it.
Step 4: Click on Start.
And, wait for few minutes to setup the SAP Build automatically for you.
Step 5: Then, click on Navigate to Subaccount.
Step 6: Once you’re inside the subaccount click on Instances and Subscriptions.
Step 7: Click on SAP Build code.
Step 8: Then, Click on Create.
Step 9 :Choose Build an Application. Then, choose SAP Build Code.
Step 10: Next, choose Full Stack Application.
Step 11: Provide a name for the project of your choice and click Create.
Step 12: Click on Ok. It will navigate you to SAP Business Application Studio.
Step 13: Click on Joule Icon.
Step 14: Type “/” in the command section and select “cap-gen-app.“
Step 15: Then, paste the below prompt and press send.
“Design a customer loyalty program application. Define 4 data entities: Customers, Products, Purchases and Redemptions. Each customer must have the following fields: name, email, 7-digit customer number, total purchase value, total reward points, total redeemed reward points. All fields for each customer should be integer except name and email that will be stored as string. Each product should have a name, description and price. Purchases should include the following fields: purchase value, reward points. All fields in Purchases must be integer. Redemptions must have 1 field in integer: redeemed amount. Each purchase and redemption will be associated to a customer. Each purchase will be associated to a product and is called selectedProduct. “
Step 16: Click on Accept .
Step 17: The screen will get updated once the prompt is executed. Then choose Open Editor. And select Sample Data.
Step 18:Choose Customers and Value 5 and press on Add. It will create 5 sample records for customer data.
Step 19: Click on Enhance.
Step 20: Next, click on StoryBoard.
Step 21: And select Purchases > Add logic.
Step 22: Click on Add.
Step 23: Then, choose on Open Code Editor > Application Logic.
Step 24: Copy the below prompt and paste it into the text field as given below press the send button. It will send the prompt to Joule to process.
“Reward points of each purchase will be the one tenth of the purchase value. Each purchase value will be added to the total purchase value of the related customer. Each reward point will be added to the total reward points of the related customer.”
Step 25: Click on Accept.
Step 26: Then, choose StoryBoard.
Step 27: And, select Create a UI application.
Step 28: Fill in the text fields as follows:
Display name: Purchases
Description: Manage Purchases
Then, click Next.
Step 29: Choose Templete-Based. And, click Next.
Step 30: Choose List report page. Click Next.
Step 31: Choose main entity as Purchases. Click on Finish.
Repeat steps 27-31 to create additional UI apps for the Customers and the Redemptions.
Details to be filled for Customer and Redemptions is given below:
Step 32: Fantastic! We’re ready to test the app. To test the application click on Run and Debug button.
A new tab opens. Click on Customers > Go.
It will open the list of customer data. Choose any one of the customers to test the edit function.
Click on the “Edit” and change Name and Email as given below, and Save.
Great, now you are able to see the list of customers and you can edit it too.
Let’s test the Create functionality of the app. Click on the Create button.
And fill the details for a new customer as given below.
The new customer is now added successfully.
Summary
In summary, this guide has detailed how to build a full-stack application from a template using SAP Build Code, incorporating SAP Joule’s Generative AI capabilities. We utilized Joule to generate code for data entities and services related to customers, purchases, and redemptions, and also to develop the application logic. By leveraging these tools, you can streamline the development process and create robust applications efficiently. Continue exploring SAP Build Code and Joule to unlock further possibilities and enhance your development projects.
Here is a detailed demo to help you learn further!
From the previous blogs you’ve learned the basics of Artificial Intelligence (AI), explored how machines can learn through Machine Learning (ML), taken a deeper dive into Deep Learning (DL), and discovered how Generative AI can create new and original content.
Now, we’re going to take it a step further by focusing on Large Language Models—a specific and revolutionary application of Generative AI that is transforming how machines understand, process, and generate human language. Let’s explore how all the concepts you’ve learned so far come together to power these cutting-edge models!
This 7 part blog series is designed for anyone who wants to start learning about SAP AI. No matter what your job is or what you do, understanding the basics of Generative AI how it connects to SAP is a smart choice.
In this blog series, we’ll break down SAP AI into easy-to-understand parts, covering one step at a time.
How Do Large Language Models Fit Into Generative AI?
Imagine you’re talking to a super smart robot, like ChatGPT, that can understand what you say and talk back to you in ways that make sense. How does it do that? Well, it uses something called a Large Language Model (LLM).
Think of the LLM like the brain of the robot. It helps the robot understand the words you say and come up with answers that sound like a human. So, whenever a robot, like ChatGPT, needs to create human-like sentences, it uses this special brain—the Large Language Model—to help it do that.
What is a Language Model?
A Large Language Model (LLM) is like a super smart robot brain that helps computers understand and talk like humans. It works by analyzing patterns in text and using statistics to predict which words are likely to come next in a sentence.
Imagine you’ve read hundreds of books, stories, and conversations, and you remember a lot about how words fit together. That’s what an LLM does! It has read tons of text from the internet, books, and other places, so it knows a lot about how people talk and write.
Here’s how it works:
Learning from Text: The LLM learns by looking at millions and millions of sentences. It sees which words are used together, how sentences are formed, and what makes sense in different situations.
Predicting the Next Word: When you type something like, “I would like to eat…” the LLM guesses what comes next, like “pizza,” “sushi,” or “ice cream,” based on all the patterns it has learned from reading.
Answering Questions: If you ask the LLM a question, like “Why is the sky blue?” it uses what it learned from reading to give you a helpful answer. It doesn’t “think” like a person, but it can find patterns in the text to give you a smart-sounding response.
Getting Better Over Time: The more text it reads, the smarter the LLM gets at predicting and understanding what people mean. It’s kind of like how you get better at reading or solving puzzles the more you practice.
So, an LLM is a super tool that helps computers understand and talk to people by learning from lots of words and sentences!
Let’s take an example to understand better!
When we use messaging apps on our phone, they help by predicting the next word as we type. For example, if you type “I would like to eat,” the phone might suggest words like “pizza,” “sushi,” or “ice cream” because it knows these are common food choices people mention after saying that phrase.
The phone makes these predictions based on the words you’ve typed and what it has learned from analyzing lots of text. It guesses what you’re likely to say next by recognizing patterns in everyday language.
LLM and NLP (Natural Language Processing)
LLMs are an advanced form of Natural Language Processing (NLP). NLP allows computers to understand and interact with humans using natural language, like English. Think of virtual assistants like Siri or Alexa, which can understand commands like, “Set an alarm for 7 AM.” NLP makes this possible by breaking down the sentence, understanding its meaning, and responding in a way that makes sense.
Where Are LLMs Used?
LLMs are used in many AI applications across industries, including:
Virtual Assistants: Siri, Alexa, and Google Assistant use LLMs to understand commands and respond accurately.
Chatbots: Models like ChatGPT help answer questions, engage in conversation, and provide support in human-like ways.
Language Translation: Tools like Google Translate use LLMs to translate text between languages while keeping the meaning intact.
Text Generation: LLMs can write articles, product descriptions, or even creative stories based on user prompts.
Summarization: They can summarize lengthy documents into short, easy-to-understand paragraphs.
Sentiment Analysis: LLMs analyze text to detect emotions, like whether a customer review is positive or negative.
Popular Large Language Models
GPT (Generative Pre-trained Transformers): Created by OpenAI, GPT is one of the most well-known LLMs, powering ChatGPT.
BERT (Bidirectional Encoder Representations from Transformers): Developed by Google, BERT helps with tasks like understanding search queries.
LaMDA (Language Model for Dialogue Applications): Another model from Google, designed specifically for conversations.
Summary
Large Language Models are the backbone of many AI applications today, making it easier for machines to understand and communicate with humans in natural language. By learning from vast amounts of text, LLMs can predict, generate, and summarize text with remarkable accuracy.
Now that we’ve covered Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning, it’s time to explore an even more fascinating area—Generative AI.
This 7 part blog series is designed for anyone who wants to start learning about SAP AI. No matter what your job is or what you do, understanding the basics of Generative AI how it connects to SAP is a smart choice.
In this blog series, we’ll break down SAP AI into easy-to-understand parts, covering one step at a time.
Generative AI is a form of artificial intelligence designed to create new content, such as artwork, music, or realistic images, without explicit instructions on what to produce.
Unlike traditional AI, which is task-oriented and problem-solving, Generative AI stands out for its ability to mimic human-like creativity.
It can generate original and unique content, ideas, or solutions, much like humans do.
Let’s Break it Down with a New Example
Imagine I asked you to design a vehicle that doesn’t exist. You’d need to use your imagination to invent something entirely new, something no one has ever seen before.
Thanks to our natural creativity, you might come up with a vehicle that has the body of a spaceship, the wheels of a monster truck, and the wings of a bird. It’s completely original, something you dreamed up from scratch.
Now, imagine a computer program doing something similar. Instead of just following a set of instructions, it can generate new things—like artwork, music, or even realistic images—without needing exact directions. It’s as if the program has its own creative spark, just like you.
Imagine a computer program has been trained by looking at countless aerial photos of eagles flying over mountains. With this knowledge, it can now generate a completely new image—not by copying any of the pictures it has seen, but by combining its understanding of what makes an eagle and a mountain unique. The result could be an entirely new perspective, like an eagle soaring high above rugged peaks, blending elements from what it learned to create something fresh and original.
Eagle Flying Over Mountains Generate by AI
This is Generative AI — a machine that can imagine and create on its own. It can draw pictures, tell stories, or even invent new games, all without needing anyone to show it how. It’s like giving a computer the ability to dream up new ideas, just like we do.
Where Does Generative AI Fit into the AI Hierarchy?
Generative AI is a branch of Deep Learning. It sits within a larger framework of artificial intelligence. Imagine AI as the broadest category, with Machine Learning being a more specific part of it. Within Machine Learning, we have Deep Learning, and finally, Generative AI is a subset of Deep Learning. The diagram below helps visualize how AI, Machine Learning, Deep Learning, and Generative AI are related.
Generative AI builds on machine learning techniques, particularly deep learning and neural networks.
What makes Generative AI stand out is its ability to create brand-new content.
Traditional AI, machine learning, and even deep learning have largely focused on predictive models—used to recognize, classify, or forecast patterns in data. For example, a typical machine learning task might be identifying a car in a set of images or grouping different vehicles based on their features.
Generative AI goes further. Instead of just identifying a car, it can create an entirely new design of a futuristic vehicle or imagine and generate something that’s never been seen before. This creative capability was once thought to be unique to humans.
The image below showcases AI’s evolution over time. The leap from rule-based systems to Generative AI has been powered by advances in learning algorithms, greater computational power, and access to vast amounts of data.
What are Generative Models?
Generative Models are the heart of Generative AI. These machine learning models learn the patterns and structures of existing data, then use that knowledge to generate new, original content.
Applications:
Visual Arts: Creating stunning images and art pieces.
Music Composition: Generating unique musical compositions.
Text Generation: Writing articles, stories, and scripts.
Common Generative Models:
Variational Autoencoders (VAEs): Known for their versatility.
Restricted Boltzmann Machines (RBMs): Often used for collaborative filtering.
Transformer-based Language Models: Excelling in natural language tasks.
We’ll explore these models in detail and see how they’re revolutionizing various industries.
Usages of Generative AI in Real-life
Here are some examples of how Generative AI is being used to create real-life applications:
Software:
Predictive Maintenance: Using AI to predict when equipment will fail, reducing downtime and costs.
Virtual Assistants: AI-powered assistants that can handle customer inquiries and tasks, improving efficiency.
Content Creation:
Creative Writing: AI can generate creative text, such as poems, scripts, or marketing copy.
Art Generation: AI can create original artwork in various styles, from abstract to realistic.
NLP(Natural Language Processing):
Language Translation: AI can translate text and speech between different languages, facilitating global communication.
Sentiment Analysis: AI can analyze text to determine the sentiment expressed (e.g., positive, negative, neutral), providing valuable insights for businesses.
Here are Some Popular Generative AI Tools
There are plenty of tools available today in the market, here are the popular ones you can try.
ChatGPT: Developed by OpenAI, ChatGPT engages in natural language conversations, generating contextually relevant and coherent responses using deep learning techniques.
AlphaCode: Created by DeepMind, AlphaCode generates optimized code solutions across various programming languages, leveraging deep learning to analyze and learn from code patterns.
GitHub Copilot: An AI-powered code completion tool by GitHub and OpenAI, integrated into code editors to provide real-time suggestions and completions.
Gemini: Google’s conversational AI chatbot, initially based on LaMDA and later upgraded to models like PaLM and Gemini.
Microsoft Copilot: An AI-powered assistant by Microsoft for tasks like summarizing articles, reformatting text, and updating images.
DALL-E: Developed by OpenAI, DALL-E generates images from text descriptions, with versions including DALL-E2 and DALL-E3.
StyleGAN: Created by NVIDIA, StyleGAN generates high-quality synthetic images, especially realistic human faces, with control over visual features.
An image generated using StyleGAN that looks like a portrait of a young woman.
Summary
In this blog, we explored what generative AI is and how it differs from other types of AI. We also discussed several real-life applications of generative AI and highlighted some popular tools in this field.
In our previous blog, we talked about how Artificial Intelligence (AI) helps a robot find your lost toy by thinking like a human. Now, we’re going to take it a step further and explore Machine Learning (ML), which is like teaching the robot how to learn on its own. Let’s dive into it in a fun and simple way!
This 7 part blog series is designed for anyone who wants to start learning about SAP AI. No matter what your job is or what you do, understanding the basics of Generative AI how it connects to SAP is a smart choice.
In this blog series, we’ll break down SAP AI into easy-to-understand parts, covering one step at a time.
In our previous blog, we discussed how AI can help a robot find your lost toy. Now, let’s explore how we can teach the robot to recognize different types of toys.
Imagine we want the robot to be able to identify various toys — like cars, dolls, and building blocks. Here’s how we can do it:
Show and Tell: We start by showing the robot lots of pictures of different types of toys. For example, we show it many pictures of cars, dolls, and building blocks, each labeled with the toy’s name.
Teach the Robot: We need to train the robot to recognize these toys based on their size, color, shape, and other features. This is like teaching a friend what your favorite toys look like so they can help you find them.
Recognize New Toys: Once the robot has seen many pictures of each type of toy, it can start to recognize them in real life, even if they don’t look exactly like the pictures. For example, even if you have a new toy car that looks a bit different from the ones we showed, the robot should still be able to identify it as a toy car.
Practice Makes Perfect: The more pictures of toys the robot sees, the better it becomes at recognizing them. This is because it learns to pick out common features and differences between toys, helping it make better guesses.
This process of teaching the robot with lots of examples is known as Machine Learning.
In summary, Machine Learning is:
A subset of Artificial Intelligence: It’s one way of making machines smart.
Learning from Examples: It involves giving machines lots of data (like pictures of toys) so they can learn and make decisions based on that data.
So, just like you learn to recognize different toys by seeing and playing with them, robots learn to identify toys by looking at many examples. The more they see, the smarter they get!
Types of Machine Learning
Now, let’s dive into the different ways robots and computers can learn from data. Machine Learning can be categorized into three main types:
Supervised Learning
Unsupervised Learning
Reinforcement Learning
Each type of Machine Learning serves a unique purpose and uses different methods to learn from data. Let’s explore each type in detail.
Supervised Learning
Remember when we taught the robot to identify various toys by showing it lots of pictures of different types of toys, each labeled with the toy’s name? This method is known as Supervised Learning.
In Supervised Learning:
We Provide the Answers: Just like you told the robot which toy is a car, doll, or block, we provide the machine with examples that are labeled with the correct answers.
The Machine Learns: The machine uses these labeled examples to learn how to identify or predict new, unlabeled data.
Here’s how Supervised Learning works in the real world:
Email Spam Filtering: Classifying emails as spam or not spam based on features from the email content and sender.
Image Classification: Identifying objects in images, such as recognizing animals or detecting objects in self-driving cars.
Facial Recognition: Verifying individuals based on facial features, used in security systems or unlocking devices.
Financial Fraud Detection: Spotting potentially fraudulent transactions by analyzing patterns in financial data.
Speech Recognition: Converting spoken language into text, like in voice assistants such as Siri or Google Assistant.
Unsupervised Learning
Let’s think about how kids learn to group their classmates on the first day of school. They notice differences and similarities among their classmates without anyone explicitly telling them how to categorize them. This is similar to Unsupervised Learning.
In Unsupervised Learning:
No Labels Provided: The machine is given data without any labels or categories. It must find patterns and group similar items on its own.
Pattern Discovery: The machine identifies similarities and differences in the data, grouping similar items together based on the patterns it discovers.
For example:
Clustering Customer Segmentation: Businesses use unsupervised learning to group customers based on purchasing behavior, helping tailor marketing efforts.
Anomaly Detection in Cybersecurity: Identifying unusual patterns in network traffic to flag potential security threats.
Recommendation Systems: Suggesting products or content based on patterns in user behavior, like how Netflix recommends movies.
Reinforcement Learning
Imagine teaching a dog a new trick. You give the dog a treat when it performs the trick correctly and no treat when it doesn’t. Over time, the dog learns to do the trick more often to get more treats. This is similar to Reinforcement Learning.
In Reinforcement Learning:
Learning by Trial and Error: An agent (like a robot or computer program) learns to make decisions by receiving rewards for good actions and penalties for bad ones.
Interactive Learning: The agent interacts with its environment, learns from the outcomes of its actions, and adjusts its behavior to maximize rewards.
For example:
Game Playing: AlphaGo, developed by DeepMind, uses reinforcement learning to play the board game Go at a superhuman level, learning the best strategies through practice.
Self-Driving Cars: Reinforcement learning helps these cars learn how to navigate traffic and make driving decisions by simulating various driving conditions.
Algorithmic Trading: Making trading decisions based on historical and real-time market data to develop effective trading strategies.
Summary
Machine Learning is a branch of Artificial Intelligence that allows computers to learn from data and experiences, without being explicitly programmed for every task. Whether it’s recognizing toys, understanding voice commands, or helping doctors analyze medical images, Machine Learning is becoming an integral part of our everyday lives.
The term “Artificial Intelligence (AI)” was first used back in 1956, but it was mostly discussed in scientific research or shown in movies. However, with the rise of ChatGPT, AI has become a hot topic everywhere.
This 7 part blog series is designed for anyone who wants to start learning about SAP AI. No matter what your job is or what you do, understanding the basics of Generative AI how it connects to SAP is a smart choice.
In this blog series, we’ll break down SAP AI into easy-to-understand parts, covering one step at a time.
Artificial Intelligence (AI) — A Kid’s Perspective
Let’s break down AI in a simple way. Imagine you’re looking for your favorite lost toy.
Here’s what you need to do to find it:
1. Identify Your Toy: You need to know exactly what your toy looks like.
2. Recognize Similar Items: If you find other toys, you need to tell if they’re the same as your lost toy. If it is the same type, you’ll check if it’s your specific toy.
3. Plan Your Search: You should come up with a plan to find your toy. For example:
First, check your room.
If it’s not there, look in the living room where you play with your toys.
If you still don’t find it, ask your family if they’ve seen it.
4. Adapt to the Situation: If your room is messy, you might need to search carefully among the clutter.
Now, imagine someone tells you, “I saw your toy in the backyard.”
Here’s what you would do:
You know where the backyard is and how to get there.
You won’t mistake a different toy for your own.
When you see the toy, you’ll check if it’s your favorite one.
You can find your toy because you have all these smart abilities.
But what if we could give these abilities to a robot so that the next time you lose your toy, the robot could help you find it?
Imagine the robot can move around and take pictures. That’s a good start, but it needs to think and act like you do. For example:
1. Identify the Room: The robot should recognize your room even if the furniture is rearranged. 2. Spot Your Toy: The robot should be able to identify your specific toy from other toys. 3. Understand Instructions: The robot needs to understand what you say and follow your directions. 4. Plan and Adapt: The robot should be able to make a search plan and adjust based on the situation, like searching more thoroughly if the room is messy.
In short, to find your toy, the robot needs to have human-like intelligence.
So Artificial Intelligence (AI) is — creating smart, human-like abilities in robots, machines, or computers.
What is AI?
Artificial Intelligence (AI) is when machines or computers are designed to imitate how humans think and make decisions.
In simple terms, AI is about making computers capable of “thinking” like humans.
AI allows computers to understand and analyze data, and make decisions on their own without needing constant human help. These smart machines use algorithms—detailed sets of instructions—to process information and get better at tasks over time.
Everyday Uses of AI
You might be using AI every day without even realizing it! Whether it’s the smart features in your apps or the technology behind your favorite services, AI is all around us, making life more convenient.
Here are some common ways AI is used in our daily lives:
Personalized Fitness Apps
Fitness apps like MyFitnessPal or Fitbit use AI to track your exercise routines and dietary habits. They analyze your data to offer customized workout plans and nutritional advice tailored to your personal goals and progress.
Navigation and Traffic Apps
When you use apps like Google Maps or Waze, AI helps find the fastest route to your destination. These apps analyze real-time traffic data to provide updates on road conditions and suggest alternative routes to avoid delays.
Smart Home Devices
Smart thermostats like Nest and security cameras like Ring use AI to learn your preferences and adjust settings automatically. For example, a smart thermostat can learn when you’re home or away and adjust the temperature to save energy while keeping you comfortable.
Email Filtering and Organization
Email services such as Gmail use AI to filter out spam and categorize your messages. AI helps organize your inbox by sorting emails into different categories, like primary, social, and promotions, so you can easily find important messages.
Streaming Music Services
Music streaming platforms like Spotify use AI to create personalized playlists based on your listening habits. AI analyzes the songs you play and suggests new music you might enjoy, enhancing your listening experience.
Voice-to-Text Apps
When you use speech-to-text features on your phone or computer, AI converts your spoken words into written text. This technology is useful for composing messages, taking notes, or creating documents hands-free.
Online Travel Booking
Travel websites like Expedia or Booking.com use AI to recommend flights and accommodations based on your past searches and preferences. AI helps you find deals and options that match your travel needs and budget.
These examples illustrate how AI is integrated into our daily routines, making tasks easier and more efficient while often working quietly behind the scenes.
Understanding AI Types
Let’s break down the types of AI in a way that’s easy to understand. Imagine AI as different kinds of robots or programs that can do different things. Here are the main types:
1. Narrow AI
Think of Narrow AI as a robot that is really good at doing just one specific job. For example, imagine a robot that can only play chess. It’s amazing at chess but can’t do anything else, like make sandwiches or tell jokes. Most of the AI we use today is Narrow AI. For instance, Siri on your phone can help with setting reminders or telling you the weather, but it can’t do your math homework or walk your dog.
Example: A calculator is a type of Narrow AI. It’s great at doing math, but it can’t help you with writing a story or making a drawing.
2. General AI
General AI is like having a super-smart robot that can do almost anything a human can do. It can play chess, help with homework, cook dinner, and even understand jokes. Right now, we don’t have General AI. It’s more like something you might see in movies where robots are almost as smart and versatile as humans.
Example: Imagine a robot that can help you with school projects, play games, and even chat with you about your favorite TV shows. This robot would be able to learn and do many different things, just like a person.
3. Artificial Superintelligence (ASI)
This is the most advanced type of AI, which is smarter than any human in every possible way. ASI would be able to solve complex problems, invent new things, and make decisions better than anyone else. But, this kind of AI only exists in science fiction for now.
Example: Think of a robot that could solve any puzzle in the world, create incredible art, and invent new technologies that we haven’t even thought of yet. It’s like having a super-genius robot who knows everything!
Generative AI fits under Narrow AI
Generative AI is a special kind of Narrow AI that can create new things. Instead of just following rules or doing tasks, it can make new content, like writing stories, drawing pictures, or composing music. For example, a Generative AI might help you write a fun story about a dragon and a spaceship, or it could generate a new piece of music based on your favorite songs.
Example: Think of an AI that helps you design a new video game level or comes up with new ideas for a storybook. It uses what it knows to create something new and original.
Generative AI fits under Narrow AI because it specializes in creating new content and isn’t yet capable of the broad range of abilities that General AI would have.
AI vs. Human Intelligence: Key Differences
Artificial Intelligence (AI) and human intelligence each have their unique strengths and characteristics. While AI is remarkable at handling numbers and rules with remarkable speed and accuracy, human intelligence brings emotional depth, creativity, and adaptability. Think of AI as a super-efficient calculator, whereas human intelligence is like a vibrant, ever-evolving work of art!
Here’s a closer look at how AI and human intelligence differ:
How They Learn:
AI: Learns by analyzing vast amounts of data and identifying patterns. It becomes proficient in specific tasks by processing numerous examples.
Humans: Learn through direct experiences, conversations, and exploration. Our learning encompasses everything from mastering new skills to understanding the beauty of natural phenomena.
Speed of Thinking:
AI: Executes tasks at lightning speed, especially those it has been trained on. It’s highly efficient with repetitive and well-defined activities.
Humans: May take a bit more time, but we excel at tackling complex problems and applying creative thinking to find innovative solutions.
Memory Capabilities:
AI: Retains information and facts based on programming but lacks personal experiences or emotions. It recalls data without the context of personal significance.
Humans: Remember events, emotions, and personal details, like cherished memories or favorite songs. Our memories are a blend of both joyful and challenging experiences.
Emotional Understanding:
AI: Does not experience emotions. It operates based on predefined rules and patterns.
Humans: Experience a wide range of emotions, such as joy, sadness, and empathy. Our feelings significantly influence our actions and interactions with others.
Adaptability:
AI: Functions based on its training and may struggle with unfamiliar situations. It is highly efficient but limited to its programmed scope.
Humans: Are highly adaptable and resourceful. We thrive in new and unpredictable scenarios, using creativity and critical thinking to navigate challenges.
Creativity and Innovation:
AI: Creates within its defined parameters. It’s like an artist working with a specific set of tools and constraints.
Humans: Are adept at generating novel ideas and solutions. Our creativity allows us to invent new concepts and push beyond existing boundaries.
Understanding Context:
AI: Knows what it has been taught but might struggle with nuanced contexts, such as interpreting humor or cultural subtleties.
Humans: Have a deep understanding of complex social interactions, humor, and cultural nuances. Our brains integrate diverse types of information to make sense of the world.
Decision-Making:
AI: Makes decisions based on data and programming. It follows logical rules and algorithms.
Humans: Use a combination of logic, emotions, and personal values to make decisions. Our choices reflect our unique perspectives and experiences.
In summary, AI and human intelligence each excel in their own ways. AI is a powerhouse of speed and precision for specific tasks, while human intelligence is enriched with creativity, emotional depth, and adaptability.
Summary
In this blog, we’ve explored the fundamental concepts of Artificial Intelligence (AI), starting from its basic definition to its real-world applications. We’ve simplified AI by comparing it to a robot that helps you find a lost toy, illustrating how AI mimics human-like problem-solving abilities. We discussed various everyday uses of AI, such as in fitness apps, navigation tools, and smart home devices, demonstrating its pervasive impact on our daily lives.
We also categorized different types of AI, including Narrow AI, which excels at specific tasks, General AI, which remains a theoretical concept with broader capabilities, and Artificial Superintelligence (ASI), a futuristic idea of AI surpassing human intelligence. Lastly, we introduced Generative AI, a type of Narrow AI that creates new content, showcasing its innovative potential.
This foundational understanding sets the stage for our upcoming discussions, where we’ll delve deeper into machine learning, deep learning, and the specifics of Generative AI. Stay tuned as we continue to break down these exciting topics in the next parts of our series.
In this blog, we’ll explore how to get started with SAP’s GenAI framework. It involves setting up of SAP AI Core and SAP AI Launchpad, and we will test the foundation model based on ChatGPT.
Let’s dive into setting up SAP AI Core and AI Launchpad
Step 1: Provision SAP AI Core in your SAP BTP Global Account.
We’ve a free tier account(not the trial). A free tier account is needed to use SAP AI Core.
Logon to SAP BTP Account and navigate to subaccount. My subaccount name is Zequance AI Trial.
Under manage assignments click on edit to add a service plan.
Then click on Add service plans.
Search “sap ai” and choose SAP AI Core. Then, choose Free Tier and Extended plan.
Click on Add 2 service plan.
Note: A free-tier account is required. SAP AI Core will not be available under plans if you’re using a 90-day trial account. Additionally, remember to save your changes after adding services.
Now under entitlements you will be able to see SAP AI Core as shown below.
Step 2: Select Boosters from the navigation pane on the left, then choose the booster for SAP AI Core.
Click on Start.
Once Checking is done, click on Next.
If you have a subaccount already, choose select subaccount.
Click on Next.
Click on Finish.
Step 3: The next step is to create keys. In the SAP BTP Cockpit, go to the Subaccounts section, select Services from the left-hand menu, and then click on Instances and Subscriptions.
To create the keys needed to access your instance, click the three dots and select Create Service Key.
Enter a service key name, for example “Booster” and click on Create.
Once your keys have been created, you can view or download them anytime by locating the key, clicking the three dots, and selecting the desired option from the available choices.
Step 4: Next, run the booster for SAP AI Launchpad. Select Boosters from the navigation pane, then choose the booster for SAP AI Launchpad from the list.
Click start.
Once checking is completed , click on Next.
Choose select subaccount and click on Next.
Click on Next.
Click on Finish.
Step 5: Click on Instances and Subscriptions from Subaccounts.
Next, click on Go to Application. It will open the SAP AI Launchpad.
Step 6: Next configure AI API Connnection.
From the AI Launchpad screen click on Add.
We need to copy and paste the keys that we have created earlier in the step 3 to AI API Connection in Launchpad as given below. Then click on Create.
Step 7: Checking for foundation-models scenario.
Expand ML Operations in left pane and click scenarios.
There you will find foundation-models.
Step 8: Creating a configuration.
Navigate to ML Operations and then select Configurations.
Provide a name for the configuration.
Select the foundation-models scenario.
Choose the version.
Select the Executable ID.
Then click Next.
In Input Parameters, give name and version of the model you want to use. We’re using GPT models. Then, click Next.
Next, click Create.
Step 9: Create a deployment. To make LLM available for use we’ve to create a virtual LLM deployment.
Once the configuration is complete, click on Create Deployment.
Choose Standard duration, and click on Review.
Click on Create.
After creating the deployment, wait until its status is updated to Running.
Step 10: Testing the models using Prompt Management functionality.
Navigate to Generative AI Hub>Prompt Management.
Now we cant Test the models in different approaches.
Text Summarization:
This example tasks the LLM to summarise a given text. We’re provided it with a long text and got a summary in 30 words as requested.
Question Answering:
In this test. we’ve asked the difference between SAP AI Core and the AI Launchpad. The answers we’ve got a perfecto!
Sentiment Analysis:
And finally we wanted to check if the LLM can identify sentiment of a text.
The LLM was asked to assess the sentiment for the below text, and it was successful in identifying the “positive” sentiment mentioned in the text.
Congrats! you’ve now successfully built an SAP Foundation Model and tested it using the AI tools SAP AI Core and SAP AI Launchpad.
You can also view this tutorial in action on our youtube channel.
In today’s digital landscape, leveraging SAP’s Business Technology Platform (BTP) can provide your business with powerful tools for Artificial Intelligence. To get started, you’ll need to create a trial account, which allows you to explore its capabilities without any financial commitment.
Here’s a step-by-step guide on how to set up your SAP BTP trial account.
Step 1: Go to www.sap.com and click on the profile icon.
Step 2: If you do not have SAP universal ID register a new account by clicking Create your SAP account.
Then a popup window appears. Enter your details and click Submit button.
Step 3: Next, you will receive an email from SAP asking you to verify your email address. Verify your email by clicking Click to activate your account.
The integration of artificial intelligence (AI) into various business operations is transforming how companies approach efficiency, accuracy, and strategic insights. From finance to supply chain management, procurement, human resources, sales, marketing, and industry-specific processes, AI is proving to be an invaluable asset. Here’s a deeper look into how AI is reshaping these key business areas.
Finance with AI
AI is enhancing financial processes by streamlining operations such as payments, expense management, financial closing, and more. By automating tedious tasks like invoice matching and auditing, AI reduces days sales outstanding and simplifies error resolution through intelligent guidance. Additionally, AI helps guard against financial fraud by detecting anomalies that might escape human notice. This not only improves financial accuracy but also boosts overall performance.
Supply Chain with AI
In supply chain management, AI facilitates a more responsive approach to customer demand and production adjustments. With AI-enabled visual inspections and predictive maintenance, companies can improve quality assurance and mitigate risks before they escalate. Intelligent auto-dispatching systems further optimize production efficiency, ensuring that resources are allocated where they are needed most.
Procurement with AI
AI transforms procurement by turning it into a proactive and data-driven operation. Automated systems provide on-screen recommendations to keep purchasing compliant and straightforward. By streamlining data extraction and minimizing manual efforts, AI reduces errors and enhances decision-making with prescriptive guidance based on historical data and trends.
Human Resources with AI
AI technology in human resources fosters a dynamic and inclusive workforce environment. It aids in personalizing career development, optimizes workforce planning through intelligent staffing analyses, and enhances talent acquisition by matching candidates with suitable job opportunities. AI also improves the employee experience by incorporating digital assistants that offer real-time help and support.
Sales and Marketing with AI
AI personalizes customer interactions by tailoring product recommendations and enhancing lead scoring to strengthen sales pipelines. By predicting customer behavior and effectively responding to it, businesses can minimize churn and maximize upselling opportunities. AI tools guide sales teams towards leads with the highest potential for conversion, optimizing sales efforts and customer engagement strategies.
Industry-Specific AI Applications
AI’s versatility extends to industry-specific applications, enabling companies to manage complexities and modernize operations. Through automation and optimized processes, businesses can stay competitive by employing predictive, data-driven strategies. This not only helps in overcoming industry challenges but also fosters innovation by enabling industry convergence and maximizing profit through AI-powered forecasting.
Looking Ahead
The application of AI across these critical sectors underscores its transformative potential in business. By automating complex processes, providing insightful data analysis, and enhancing decision-making capabilities, AI is not just a tool for innovation—it is a foundational element that propels businesses towards a more efficient, accurate, and insightful future. As industries continue to embrace AI, the scope for growth and improvement expands, paving the way for a new era of business operations that are smarter, more responsive, and continuously evolving.
SAP’s approach to AI goes beyond the usual. SAP said it will spend 2 billion euros on AI, which is huge. SAP doesn’t see AI just as an add-on but as a key part of its business software, smoothly built into it.
This integration of GenAI technology with Business Applications is more than just an improvement; it marks a major change towards smarter business processes, ready to change the way companies work, make decisions, and enhance their performance.
The Dawn of Generative AI in Business
SAP’s foray into generative AI is driven by a vision to deliver unmatched value to its customers. By contextualizing generative AI within business applications, SAP aims to unlock the potential that was once deemed unattainable. The focus is on creating tangible benefits that resonate across various industries.
In a series of recent announcements, SAP has outlined its roadmap for generative AI and its commitment to innovation. The launch of a generative AI hub at the end of 2023, and the introduction of the SAP HANA Cloud Vector Engine, expected by March 2024, are testament to SAP’s strategic investment in AI technology.
Joule: The Next-Generation Digital Assistant
A standout innovation in SAP’s AI portfolio is Joule, it was launched in Sep 2023, a generative AI-powered digital assistant. Joule represents a leap forward in how businesses interact with AI, offering personalized assistance for daily tasks.
Joule is embedded with SAP’s enterprise cloud portfolio. It is not merely an addition to the existing tools but a transformative natural-language, generative AI copilot designed to revolutionize the efficiency and effectiveness of businesses across various sizes and sectors. This innovation promises to integrate seamlessly into daily operations, leveraging the power of artificial intelligence.
SAP’s Generative AI Architecture: A Blueprint for Innovation
At the heart of SAP’s generative AI strategy lies a deep understanding of business processes, which guides the integration of AI across its application portfolio.
The architecture below shows how Generative AI is placed through SAP’s Cloud portfolio. It comes with the tools and libraries that you need to build and customize the solution.
SAP is developing a wide range of use cases from question answering, text generation, classification, and exploring the potential of emerging paradigms such as LLM agents.
Looking Ahead: SAP’s Vision for the Future
SAP’s vision for the future of generative AI in business applications is both ambitious and forward-thinking. The company plans to continue enhancing its platform, integrating generative AI in innovative ways that go beyond current applications. This includes the development of writing assistants, intelligent process recommendations, and other AI-driven tools designed to empower businesses. SAP’s roadmap for generative AI reflects a commitment to staying at the forefront of technology, driving progress, and delivering value in the ever-evolving digital landscape.
In conclusion, SAP’s integrated approach to generative AI is not just a strategic move; it’s a vision for the future of business technology. By embedding AI into the fabric of business applications, SAP is setting a new standard for intelligence, efficiency, and innovation. As companies navigate the complexities of digital transformation, SAP’s generative AI solutions stand as a beacon of possibility, offering a glimpse into a future where businesses are smarter, more agile, and more connected than ever before.
