How often have you had sales tell you about a company’s ‘transformational’ enterprise AI technology, only to sit through hours of presentations and architecture diagrams without actually experiencing the technology? How often have you wondered if the company is hiding their technology from you as they drone on about their slides? This is an experience that can leave you more confused than confident. Unfortunately, this is all too common in enterprise AI.
The arrival of ChatGPT sparked a frenzy for bringing AI agents into the enterprise. It captured the imagination of everyone from geeks to governments. In hindsight, it’s not surprising that it became the fastest application to get to 100 million users in the history of technology products that had seen viral products before like Facebook, Instagram, WhatsApp, and Gmail.
The importance of LLMs as a technology was not lost on CEOs. Operating in an environment of rapidly increasing interest rates and slowing economic growth, they all urged their technology leaders and CIOs “bring ChatGPT to the enterprise” to drive business productivity.
The Cambrian explosion from an app that could write a poem to a paradigm-shifting business technology was best captured by a cartoon.
Image from Tom Fishburne
The ChatGPT feeding frenzy
On surface, this looked like a trivial request but underneath, the ask was very complex: Improve business operations with the power of LLMs, and deliver a ChatGPT-like experience for employees by connecting it to data and workflows across disparate business systems, breaking application silos and human-aided workflows that decrease business efficiency. Turns out, the generative aspect of LLMs aren't enough to make this happen. You need a system that's capable of reasoning, similar to the way that humans do.
At Moveworks, we saw the future over a year ago and bet on building an architecture of the future, one powered by a sophisticated Reasoning Engine which would help to deliver against that vision.
Skeptical and disappointed tech buyers
The feeding frenzy was not lost on vendors, who repainted their websites and added “AI” to all of their marketing collateral, with nearly identical messaging. This AI washing by vendors left even seasoned tech buyers confused and skeptical. There were two reasons for this:
Vendors refused to let buyers try their AI products without elaborate sales cycles. Even then, they ensured buyers followed tightly scripted demos. Buyers couldn’t tell what was real and what was a scripted demo, or just an old school decision tree-based chatbot.
What made problems worse was that internally developed POCs and demos also left a lot to be desired. They couldn’t imagine how an AI system could understand their business context, data, and APIs for finding information and taking action.
These challenges prevent technology buyers from figuring out what’s real and what’s not. As a result, they are still unable to deploy generative AI agents in their enterprise, and their businesses suffer as a result. This slows down the much-needed impetus for delivering business productivity.
Designing with empathy for enterprise buyers
Late last year, we decided that we would go against the grain of enterprise software vendors, and open up our platform for the entire world to experience firsthand. We were motivated by the desire to empower our buyers to experience what an agentic AI assistant with powerful reasoning feels like
Introducing Moveworks Playground
Moveworks Playground is your go-to destination to experience the power of AI agents in the enterprise — without any pressure or commitment (let alone talking to anyone from sales!)
What is the Moveworks Playground?
It’s a freely accessible version of our agentic AI assistant, connected to our Reasoning Engine where you are at liberty to explore a variety of prompts and use cases. To eliminate friction in getting to the core experience, we’ve provided sample data and actions - that are representative of your business data and automations. This allows you to experience what it feels like to have a powerful AI assistant that knows you, your organization, your business systems, and helps you find information, take action, generate insights, and get things done with real AI, today.
Who is the Playground for?
We’ve designed the Playground for anyone who wants to experience the power of AI agents in the enterprise firsthand. You don’t have to be tech-savvy to explore this.
What can you do with the Playground?
You can more or less type or prompt anything you want as long as you remember these core tenets:
It uses the sample data and actions to assist you. Remember actions are only examples and won’t modify underlying data.
It will not use general world knowledge to answer questions. It is designed to adhere to the sample data and actions, as it would if you connected it to your business data and systems.
Beyond these guidelines, we encourage you to try many variations of prompts. For example:
Simple requests for quick info:
- How much time off can I take as a new hire?
- The printer's jammed—what should I do?
Quick tasks for handling automated actions:
- Can you book next Thursday off for me?
- Can you reset my password?
- I got invited to a customer call on Zoom—can you get me access?
Complex queries to answer multiple questions:
- Can you compare our vacation policies in the US and India and tell me which is more generous?
- Does Stark Industries or Wayne Enterprises have more paid users? What's the diff?
Planning-oriented instructions that pull info from one source to use in another:
- If my customer lunch expense is still pending, can you file a ticket to expedite this?
Whether your query is simple or multi-part, Playground is able to handle it. So why not skip the sales call, and try out the power of AI agents in the enterprise for yourself in the Moveworks Playground.
Table of contents
