Generative AI, or artificial intelligence, has the power to change how we live and work in so many ways; our creativity is the only limit.

At a recent roundtable discussion with Qlik entitled The Future of Data Analytics in the Age of Generative AI, I shared my thoughts about how the newly released foundation models in language or Large language models (LLMs) as we call them today are reshaping the work landscape. LLMs like GPT4, Claude etc., have been fine-tuned using reinforcement learning with human feedback to enable different categories of uses case, as listed below:

LLM as a language facade

Using LLMs as a layer of communication between humans and machines will make talking to software as easy as talking to another person. Instead of clicking through menus, you would just tell the software what you want and facilitate a seamless flow of information, transforming the way we interact with technology.

LLM as a co-pilot

Envision your digital sidekick enhancing your productivity exponentially, a testament to the possibilities of generative AI. LLMs could aid software programmers, supercharging their efficiency. Similarly, artists could use these models for inspiration, discovering new and creative ideas they haven’t thought of before.

LLM as a role-player

The intricate world model these LLMs acquire through training with trillions of words enables them to don any role and act in character. The power of role-play is limited only by our imagination — they could be coaches, companions, or even therapists.

LLM as an orchestrator

Moving beyond single-step interaction, LLMs could handle a series of tasks, making abstract interactions more concrete. Imagine your digital personal assistant breaking down complex processes into sub-tasks and diligently completing them step by step.

Also Read: How to stay creative in the age of Generative AI and Web3

While we are only at the inception of this transformation, several early use cases have already started to gain traction:

Q&A bot based on a data corpus

In the current search engine paradigm, information synthesis is a manual process. Generative AI models like ChatGPT bridges this gap, internalizing and summarising vast amounts of data, offering succinct and accurate responses. The technology eliminates redundant research, enabling users to devote their time to higher-value tasks. We can also engineer them to cite the sources of information they present to ground them.

Customer service bot

AI models can now imitate a range of communication styles and interact with customers in unique ways, allowing for the customisation of content and its delivery. They can change how they talk, their empathy, and their style based on how they want to talk to the customer.

This allows us to not only personalise what we tell the customer but also how we say it. The depth and reasoning behind every response can be engineered, taking customer service to an unprecedented level.

Coaching

AI coaching assistants can now provide an intelligent, interactive training experience. Be it for sales forces or for children learning new concepts, the models can role-play, ask follow-up questions, and provide feedback, offering a personalised learning experience.

In sales training, the AI bot can pretend to be a customer and ask good follow-up questions, pushing trainees to think about what the customer needs and how to sell to them. In education, these bots can act as personal tutors for kids, helping them understand what they’re learning.

Only the beginning

The power of generative AI lies in its ability to democratise data, bringing unstructured and structured data together to unlock business value in enterprises.

We are actively collaborating with companies like Qlik, for example, to see how we might augment and automate manual tasks involved with data management in order to help companies boost their productivity with quality and governance in mind.

Imagination and innovation would carry us to a future of work that we envisage. The key value of AI is in human augmentation – shifting employees and human labour to higher-value work.

However, a lot of engineering still needs to be done to put safeguards in place to make this technology more robust, safe, and fit for customer interaction.

You can watch my full presentation here.

—

Editor’s note: e27 aims to foster thought leadership by publishing views from the community. Share your opinion by submitting an article, video, podcast, or infographic

Join our e27 Telegram group, FB community, or like the e27 Facebook page

Image credit: Canva Pro

This article was first published on June 7, 2023

The post How to unlock new horizons with generative AI appeared first on e27.