Differences Between GenAI and AI

Generative AI (GenAI) is a hot topic right now, but our customers have been leveraging AI for use cases such as customer churn, fraud detection, recommendation engines, and personalization for many years. What is the difference between GenAI and Analytical AI?

GenAI are the large language models that are creating new content by taking a prompt as input and generating new outputs. For example, it can generate images, poems, or in the case of automation and automated assistance, help to generate code. The GenAI models tend to be, by default, fairly generic and trained on a very wide variety of information available on the internet.

Analytical AI is used to classify, group, or predict things by leveraging tuned models that are very specific to a business' data and use cases. Analytical AI is typically applied for very pointed solutions. Many of the cases are common enough and happen in a standard enough way that they have been packaged up into an AWS service. Fraud detection, text to speech, speech to text - these are some of the many forms of artificial intelligence or machine learning that are available to consume without the effort of training and maintaining the underlying models. 

Custom models built from company data with proprietary information can address company specific problems for use cases like fraud prevention. Companies like Mastercard, Visa, and American Express are very interested in fraud detection, not only to prevent it, but also to avoid any non-fraudulent transactions from being declined unnecessarily.

When thinking about GenAI, this changes because the large language models that have been trained to date have effectively been trained on the public internet. The result is that they're very generalized across all the corpus of human information that's on the internet and they're less specific to a particular business' use case. Using large language models out of the box may not allow a company to differentiate itself from their competitors, but could raise the bar for productivity.

The key cases we see today for GenAI are used to accelerate internal use cases. For example, organizations may use this for generating content or summarizing data. 

Another main approach that we are seeing companies using is Retrieval Augmented Generation (RAG) in combination with GenAI. Companies can integrate their proprietary information, ideally privately using services like Amazon Bedrock or Amazon SageMaker JumpStart, with these models to generate a natural language response. They can then use a conversational interface to find answers from their internal knowledge and generate summarized responses.

For example, it can help an account executive draft a message to respond to an RFP for a customer, based on their internal documentation. Using the internal corpus of company data to help generate that response is still very specific to the internal company use case but leverages the language capabilities that the large language models have been trained on.

The key is not to fully retrain the large language models, but to mix in the optimal amount of specific tuning in order to see success. These models aren’t sentient and they don’t have intent, they just know the data that has been fed into the model. We can use this to our advantage, in order to leverage its ability to understand and produce language and outputs combined with company data to get a tuned response for business use cases.

Although AI services such as Amazon Forecast, Amazon Transcribe, Amazon Textract, Amazon Lex, and Amazon Polly aren’t used for GenAI, AWS has two key services that are important for GenAI. These services are Amazon SageMaker JumpStart and Amazon Bedrock. Both of these services are model hosting options that can be used with third party models that have already been trained across this huge amount of data. A company’s data isn't being used to retrain the model, so it can be used internally for competitive differentiators for an organization’s own use cases.

To summarize, GenAI harnesses large language models to generate new content or responses based on textual prompts, offering a broader and more generalized approach. It's particularly useful for internal use cases like content generation and data summarization, while AI typically involves more specific, business-tailored models for classification, detection, or prediction tasks such as fraud detection or text-to-speech conversion. 

___________________

Are you exploring ways to take advantage of Analytical or Generative AI in your organization? Partnered with AWS, Caylent's data engineers have been implementing AI solutions extensively and are also helping businesses develop AI strategies that will generate real ROI. For some examples, take a look at our Generative AI offerings.