Industrial Application of Generative AI

The year 2030, Marvel Universe released Avengers Reboot, a massive blockbuster movie, 90% of the film was generated by AI. This is not my figment of imagination, but a prediction by Gartner!

ChatGPT is an artificial intelligence (AI) program designed to communicate with people using natural language. It is based on a complex deep-learning neural network architecture that allows it to understand and respond to text-based queries in a human-like manner.

At a high level:

• At the core of ChatGPT is a deep neural network architecture that has been pre-trained on vast amounts of text data from the internet.
• The network is composed of many layers of artificial neurons, each of which processes a different aspect of the input text and feeds its output to the next layer.
• The model uses a technique called “transformer” to learn contextual relationships between words and phrases in the input text, which helps it generate more accurate and relevant responses.
• When a user inputs a query or question, ChatGPT processes the text through its neural network architecture and generates a response based on its understanding of the query and the context in which it was presented.
• The response is generated using a combination of language models and probabilistic methods, which allow ChatGPT to generate fluent and coherent text that is relevant to the input query.

The basic working of ChatGPT is depicted below:

1. Healthcare: Drug Design

According to a study conducted in 2010, the process of bringing a new drug to market was found to cost an average of $1.8 billion, with drug discovery accounting for approximately one-third of that amount. Moreover, the discovery process alone took a significant amount of time, lasting anywhere from three to six years. However, the use of generative AI has enabled the rapid design of drugs for different applications in a matter of months, presenting pharmaceutical companies with the potential to significantly reduce both the costs and timeline associated with drug discovery.

2. Manufacturing: Semiconductor

Generative AI has the capability to use a machine learning technique known as reinforcement learning, which can be applied to optimize the placement of components in semiconductor chip design, also known as floor planning. This involves determining the optimal placement of circuit components on a chip to minimize signal interference and ensure that the chip operates efficiently.

Traditionally, this task required a team of experienced human experts and could take weeks to complete. However, with the use of generative AI, the time required to complete the process can be reduced significantly to just a few hours.

The generative AI system is trained on a large dataset of previously designed semiconductor chips, enabling it to learn the best practices for floor planning. It can then apply this knowledge to new designs, rapidly generating optimal floorplans by continually refining its approach through reinforcement learning.

The use of generative AI in this context offers significant advantages, including faster product-development life cycle times, improved chip performance, and increased efficiency. It also frees up human experts to focus on more complex tasks and design challenges, improving overall productivity and innovation in the semiconductor industry.

3. Manufacturing: Supplier Assessment/Management

Finding the most suitable supplier for a specific program can be a challenging task, and sourcing teams often have to sift through an extensive range of supplier capability documents, performance data, and a broad RFX process to identify potential vendors. Typically, supplier capability profiles exist as raw data in the form of pdfs, ppts, or Word documents across multiple sources. The process of identifying the optimal supplier for a given program can be highly manual and labor-intensive.

However, the use of generative AI in combination with predictive models can significantly accelerate the vendor selection process for a particular program. The generative AI model utilizes past performance data from vendors, as well as vendor data from both public and paid sources, to build a risk value profile. This profile can then suggest the top vendors for the program, enabling the sourcing team to streamline the selection process and reduce the amount of time and effort required to identify potential suppliers.

Overall, the use of generative AI and predictive models offer a valuable tool for sourcing teams, helping them to make more informed decisions and identify the most suitable vendors more efficiently.

4. Retail: Conversational search

Conversational commerce powered by generative AI involves the use of natural language processing (NLP) and machine learning algorithms to interpret customer requests and provide relevant responses. This requires a significant amount of data processing and analysis, including the development of models trained on vast amounts of data to ensure accurate and effective responses.

The implementation of “chatGPT-like” search bars involves the development of custom NLP models trained on retailer-specific data sets, enabling the system to understand customer intent and provide tailored product recommendations. This approach requires a deep understanding of the retailer’s product catalog and customer behavior, as well as the ability to integrate with existing e-commerce platforms and technologies.

Overall, conversational commerce powered by generative AI involves complex technical processes, including the development and training of sophisticated NLP models and the integration with existing e-commerce systems. However, the potential benefits in terms of improved customer experience and increased sales make it a valuable investment for retailers looking to stay competitive in the digital marketplace.

5. Retail: Customer Service

The use of ChatGPT can enable retailers to offer immediate responses to customer queries, including product details, store locations, and shipping information. This approach can enhance customer satisfaction and alleviate the workload of customer service representatives, resulting in greater efficiency and improved customer service.

From healthcare and manufacturing to retail, generative AI has proven to be a game-changer, revolutionizing processes and enhancing efficiency. The rapid design of drugs, optimization of semiconductor chip designs, streamlined supplier assessment, conversational search, and improved customer service are just a few examples of how generative AI is reshaping industries.

With its ability to generate accurate and relevant responses, generative AI, such as ChatGPT, empowers businesses to make informed decisions, reduce costs, accelerate processes, and ultimately provide better experiences for their customers. As we look towards the future, it is evident that generative AI will continue to play a pivotal role in transforming industries and driving innovation forward.