Published on December 9, 2023, 11:11 am

Generative Ai: Reshaping Enterprise Applications For The Future

Generative AI: Revolutionizing Enterprise Applications

Generative AI is ushering in a new era of technology, poised to reshape the business landscape for years to come. The impact of this transformative technology is already being felt, as evidenced by the growing investments and interest in generative AI. While it may seem like a rapid transformation, integrating generative AI into enterprise systems is a steady and ongoing process that requires careful implementation.

The initial focus lies within the infrastructure layer, where companies are investing in the necessary building blocks for power and performance. Capital is flooding into companies like Nvidia and GPU aggregators, signaling the progress being made in this domain. However, as adoption and investment move up the stack, attention will shift towards developing new experiences and products that will redefine subsequent layers.

We are just beginning to witness how generative AI will revolutionize applications at the surface level. Even before generative AI, enterprise applications were evolving to deliver more user-friendly experiences by improving UIs and incorporating interactive elements. This shift led to a transition from “system of record” applications to “system of engagement” applications. Collaboration became the defining characteristic of these new tools, with features like multiplayer mode, annotation functionality, version history, and metadata enabling seamless content sharing within organizations.

As generative AI shapes the future of application development, we can expect even more significant advancements. Initially, we’ve seen lightweight tools emerge that leverage chat-based models such as ChatGPT. These tools provide immediate but limited value and often result in high churn rates due to lack of workflow integration or additional functionalities. Moreover, these applications generate single-use content without embedding it into users’ everyday workflows.

The second wave of generative AI applications is now taking shape. These applications aim to integrate structured data from system-of-record applications with unstructured data from system-of-engagement applications using generative models. Companies tackling this challenge have a higher chance of building enduring businesses compared to first-wave entrants. However, they must find a way to build an “ownable” layer above the system-of-engagement and system-of-record applications, even as incumbents like Salesforce scramble to implement generative AI.

This leads us to the third wave where new players establish their own defensible “system of intelligence” layer. Startups will begin by introducing innovative offerings that leverage existing system-of-record and system-of-engagement capabilities, ultimately building workflows that can be standalone enterprise applications. Rather than replacing interactive or database layers, these players will create new structured and unstructured data. Generative models will use this data to enhance product experiences, resulting in the creation of “super datasets.”

Key areas of focus for these products should include integration capabilities for data ingestion, cleaning, and labeling. For example, while building a new customer support experience, ingesting existing customer support tickets is not enough. A truly compelling product should also incorporate bug tracking, product documentation, internal team communications, and more. Insights will be derived from relevant information extraction, tagging, and weighing to provide novel insights. With feedback loops for training and usage within organizations and across multiple organizations, these products become difficult to replace due to the highly valuable weighted and cleaned data they possess.

At this stage, intelligence lies not only in the product or model itself but also in associated hierarchies, labels, and weights. These true system-of-intelligence products leverage generative AI capabilities to deliver insights in minutes instead of days. The focus shifts from synthesizing information to actions and decisions based on that information.

When it comes to endurance in a highly competitive market landscape, the least important product often gets eliminated first when budgets tighten. Emerging system-of-intelligence products need to provide enduring value to survive against both immense competition and established incumbents who are incorporating generative AI into their offerings. This requires coupling high-value workflows with collaboration features and introducing super datasets into their platforms.

The AI space has been rapidly evolving over the past year. Open source models are proliferating, while closed proprietary models are also advancing at an exceptional pace. Now, it is up to founders and innovators to build enduring system-of-intelligence products on top of this shifting landscape. When done right, the impact of these products on enterprises will be extraordinary and long-lasting. Generative AI is not just a buzzword; it truly has the potential to transform how businesses operate in the future.


Comments are closed.