The Next Evolution: Generative AI to Interactive AI

The Rise of Generative AI

Generative AI captured the world’s imagination because the pattern of interaction is similar to human interaction. Ask a question and get an answer of varying accuracy and relevance. The quality of the answer depends on who or which Gen AI model you ask, what you ask and how you ask. With humans this is thought of as quality of communication. In the Generative AI space this is called prompt engineering.

The attraction of Generative AI both to consumers and businesses is both volume of data the foundational models have been trained on and the speed of response. For the business enterprise, Generative AI accelerates time to value of AI as the foundational models are pre-trained on more data than 99% of companies can afford to train their AI models on. They provide an excellent starting point for AI initiatives. This has not gone un-noticed by the hyper-scalers, large technology companies and the venture capital community which has made nearly unprecedented levels of investment in Generative AI. What this means is where we are today is not where we will be tomorrow.

The Evolutionary History of AI

Beginning with the advent of computers in the enterprise statistical analysis evolved into business intelligence which evolved into advanced analytics which evolved into “early stage AI”. This is where we were in the early 2000s with more advanced AI techniques that are common today, essentially living in academia or in highly specialized areas. Beginning in the early 2000s, compute power and data storage increased in capability and decreased in cost at the same time as broadband internet and smart phones became popular. Visionary leaders and teams at Amazon, Apple, Facebook, Netflix and Google were some of the earliest companies to capitalize on the business opportunity the confluence of these events represented.

At the same time, the open source movement spearheaded by Red Hat, followed by the Hadoop ISVs, Confluence and Databricks began to gain momentum and there was more sharing on advanced frameworks and methods in the Big Data and AI community empowering and enabling greater variety and velocity of innovation. B2B and B2C applications that leveraged Natural Language Processing (NLP) and Computer Vision became more common.? These applications were the precursors to Generative AI.

The Next Evolution

The conversational pattern of ask a question, get an answer is the simplest form of human interaction. The next level up is one person asking another to do something, getting an affirmative response and having a high degree of confidence that the task will be completed as it should be on a timely basis. Interactive AI will reflect this pattern.

Bing & Google as representative internet search engines represented an early maturity model of AI/human interaction that was the precursor to Generative AI. For example, ask your search engine of choice “Italian restaurant near me” and you will get a set of reasonable recommendations. Today’s voice assistants Alexa/Google Voice/Siri are low maturity models of Interactive AI. Ask your voice assistant “Play Taylor Swift’s greatest hits” and you have your Taylor fix for the day.

Generative AI is the precursor to Interactive AI. Interactive AI will respond to user prompts and act as an agent on your behalf to accomplish multi-step tasks. For example, “list the item I just took a photo of on eBay” or “take June’s PowerPoint financial report, pull July’s financials and put into our standard monthly PowerPoint format” or as we demonstrated recently with Cognizant’s Neuro AI, we can leverage a properly trained set of Gen AI models for complex tasks like designing a telecommunications network.

How to Prepare

Chances are if you are reading this the thought occurs “We are early in our Gen AI journey, how can we even start thinking about Interactive AI?” At Cognizant, we believe every company needs a well considered fit-for-purpose AI strategy that provides the foundation and framework to efficiently and effectively make business and technology decisions that are aligned to your mission. In both the AI and technology space the rate of change of the rate of change will continue to accelerate. The skillful response is to recognize this, accept it and to be prepared for the inevitable surprises that both today and years from now will make the saying “may you live in interesting times” more and more true every day.

?

?

?

?