This tool can significantly improve legal professionals’ workflows by enabling them to analyze large amounts of text quickly, enhance accuracy by identifying inconsistencies, proactively manage risks, and inform negotiation strategies.
For our example we have chosen two different NDAs and to make it more illustrative, we compare the term and penalty clauses. However, you can use different contracts or clauses to streamline your own workflow.
First, your client or colleague needs to upload the desired contract in PDF format and select the topic the want to compare. Then, click on Next, and within a few seconds, they will receive the comparison they were looking for.
– Let’s build this bot step by step! –
Step 1: Setting up your bot
As always, the first thing we need to do is to type a concise name and a brief description for our bot.
Next, include a files node. Make sure to toggle on “use in AI output” so that the AI knows the document uploaded in the front end is the one to be compared Also, choose an existing AI setting or create a new one tailored to your bot’s requirements. In this example, we have set up a specific AI configuration. We carefully selected the necessary parameters for the AI to vectorise the NDA provided on the frontend.
Right below, we add a dropdown node from which the client can choose the specific topic they want to compare.
Step 2: AI Engine
And here comes the core functionality of our bot, the AI Output node.
Here, we need to select an AI settings or create a new setting to ensure our AI operates according to the desired quality and type of response. A crucial aspect of the AI setting for this AI output node is uploading a master template as the knowledge base.
In this example, we have created a new AI setting and we uploaded an NDA template. This allows the AI to compare every other NDA uploaded on the frontend with the one provided in the knowledge base. Next, we can type some default values for before and while the AI is processing the information. Now, we need to connect the AI Output node to the previous files node. Activate the switch and choose the Files node name. By doing this, we grant the AI access to the document uploaded on the frontend.
Now it’s time for prompt creation, a step you’re likely familiar with by now. In the system prompt, we have provided instructions for the AI to compare an uploaded contract with a defined market standard and the template in the knowledge base. The AI should also briefly describe the differences and output the results in HTML. In the dynamic prompt, we have defined the query, which in this case is the topic to be compared, selected from the dropdown in the frontend. To do this, we insert the dropdown variable.
To finalize this node, we have chosen to trigger the AI response when the node is being loaded, but your bot may require a different trigger option.
Afterwards, with the purpose of making the AI result visible on the front end, we include a text field node inserting the ai output variable.
Step 3: Storage your data
Finally, with the aim of saving the input in the database and making it accessible in your bot’s statistics dashboard, include a “send and result” node.
And just like that, gone are the days of tedious manual contract comparisons!
With the magic of this AI-powered tool, you can effortlessly analyze every relevant contract in a fraction of the time, ensuring you have a clear understanding of the terms ensuring you have a clear understanding of the terms and maximizing your productivity.
We hope you found this tutorial helpful and that it inspires you to develop your own contract comparison bots or incorporate this functionality into more complex bots.
Happy automating with e!
