Using AI in Casebase.

What you can expect

How does artificial intelligence (AI) work?

What are the current functions of AI in Casebase?

What are the current limitations of AI in Casebase?

How to improve and use AI models in Casebase?

How does artificial intelligence (AI) work?

Artificial intelligence (AI) denotes human-developed technology that systematically employs extensive datasets within an iterative framework. This enables the AI to forecast outcomes, generating mathematical or linguistic outputs in response to user inputs.

• The AI discerns patterns within an extensive dataset, known as training data, to construct a model.
• Subsequently, the model’s accuracy is assessed through posing a question for which the AI already possesses the answer, and its response is analyzed.
• The data produced by the AI during this process is termed as test data.

As time progresses and more inputs and test data are integrated, the AI refines and enhances its model through iterative learning.

Note

Artificial intelligence is an extremely complex science made up of machine learning, neural networks, cognitive computing, and so much more. While we aim to give you practical context behind Casebase new AI features, this guide is not a comprehensive resource to learn about artificial intelligence.

If you would like basic training in data literacy or AI/ML, contact us or our colleagues from [at]. 

What are the current functions of AI in Casebase?

Currently we offer two functions based on a Larg Language model (LLM). They should help you in your workflow.

Brainstorm new use case ideas — sometimes inspiration strikes right when we need it. We have implemented our AI use cases assisstant which helps you to find the right lid on the pot. Simply describe your business problems in a few short sentences in the submit template. Then “Ask AI” and the system will suggest you three ideas how you could solve your business problem with a concrete Data Analytics or AI/ML use case. This will be a great source of inspiration for you.

Ask our chatbot everything about the EU AI Act — You have certainly heard of the EU AI Act! But what exactly is behind it and which paragraphs bring which requirements with them is maybe unclear to you? To be honest, you don’t need to know that either. Because we have integrated a chatbot in our AI Act Risk Check feature that you can ask everything you want to know about the AI Act. So regulation topics are only half as bad. 

What are the current limitations of AI in Casebase?

All AI outputs are limited and AI continues to be an area of rapid research and development. Limitations in AI as a whole can be attributed to many different causes, including but not limited to: narrow inputs, systemic biases, limited quantity of training or test data, or poor quality of training or test data.

As a consequence, AI features in Casebase:

• may output incorrect information.
• may output harmful content when prompted.
• may not know about recent events occurring in the last 6-12 months and may output inaccurate or outdated results.
• may contain bias and output biased responses.

Artificial intelligence and machine learning models can improve over time to better address specific use cases. AI in Casebase is the same — the feature will continue to get better in the coming months.

What about privacy?

Your data is still encrypted and protected according to our standard privacy practices.

The AI features in Casebase do not use your data to train our models. Any information used for AI in Casebase is shared with our partners for the sole purpose of providing Casebase’s AI features to you. We do not allow our partners or third-party vendors to use your data to train their models or for any other purpose.

How to improve and use AI models in Casebase?

Improving AI in Casebase. Casebase does not use your Customer Data or permit others to use your Customer Data to train the machine learning models used to provide our AI features. Your use of the AI features does not grant us any right or license to your Customer Data to train our machine learning models.

Artificial intelligence and machine learning models can improve over time to better address specific use cases. We may use data we collect from your use of our AI in Casebase features to improve our models when you (i) voluntarily provide Feedback to us such as by labeling Output with a thumbs up or thumbs down; or (ii) give us your permission.

Input and Output. You may provide input to be processed by Casebase (“Input”), and receive output generated and returned by Casebase based on the Input (“Output”). When you use Casebase, Input and Output are your Customer Data. You are solely responsible for the development, content, operation, maintenance, and use of your Customer Data. You will ensure that your Input and use of our Casebase AI features and Output will not violate any applicable law; or misappropriate any of our rights or the rights of any third party. You acknowledge that due to the nature of machine learning and the technology powering AI features in Casebase features, Output may not be unique and Casebase may generate the same or similar output to  a third party.

Casebase AI Feature Use Restrictions. You may not use AI in Casebase or Output (i) to develop foundation models or other large scale models that compete with Casebase; (ii) to mislead any person that Output from the Services was solely human generated.

WARRANTY DISCLAIMER. CASEBASE DOES NOT MAKE ANY WARRANTY AS TO THE RESULTS THAT MAY BE OBTAINED FROM THE USE OF AI IN CASEBASE OR THE ACCURACY OF ANY OTHER INFORMATION OBTAINED THROUGH AI IN CASEBASE. YOU UNDERSTAND AND AGREE THAT ANY MATERIAL AND/OR DATA OBTAINED THROUGH THE USE OF ANY AI FEATURE IN CASEBASE IS DONE AT YOUR SOLE RISK. YOU SHOULD NOT RELY ON FACTUAL ASSERTIONS IN OUTPUT WITHOUT INDEPENDENTLY FACT CHECKING THEIR ACCURACY. OUTPUT THAT APPEAR ACCURATE BECAUSE OF THEIR DETAIL OF SPECIFICITY MAY STILL CONTAIN MATERIAL INACCURACIES. CASBASE CANNOT DYNAMICALLY RETRIEVE INFORMATION, AND OUTPUT MAY NOT ACCOUNT FOR EVENTS OR CHANGES TO UNDERLYING FACTS OCCURRING AFTER THE AI MODEL WAS TRAINED. NO INFORMATION OR ADVICE, WHETHER ORAL OR WRITTEN, OBTAINED BY YOU FROM CASEBASE SHALL CREATE ANY WARRANTY NOT EXPRESSLY MADE HEREIN.

Service Level Terms Are Not Applicable.  Notwithstanding anything to the contrary in your Agreement or the Service Level Terms, downtime of AI in Casebase that results from a failure of a third party service will not be included in the Availability and Downtime calculations.

Third Party Provider Policies. If you choose to use the Casebase AI feature(s), you may not use the AI features in a manner that violates any OpenAI Policy, including their Content Policy; Sharing and Publication Policy; and Community Guidelines.