From personal helper to firm-wide practice
Today, you’ll see how to take great AI use cases and make it into a simple tool or agent that anyone in your team can use.
We will look at options in ChatGPT and in Microsoft Copilot.
We will also look at the process I like going through to implement this with your team.
Real-World Use Cases to Spark Ideas
Here are three real-life examples I worked on in different industries:
Referral intake assistant for clinics
An optometry clinic receives a referral email from a GP.
The AI agent finds all the details about the patient and the case, so the team can easily put it into their system.
It asks a few questions and then drafts the referral acknowledgement to send back to the GP.
Consent order helper for lawyers
A lawyer looks at back and forth emails around a parenting schedule.
The AI agent works through its template to find all the details it can, shows the layer what it's found and asks for what's missing.
Then it uses its template to draft a consent order.
Entity establishment assistant for accountants
An accountant wants to set up a company or a trust for their client.
The AI agent finds all the required information in email threads, call transcripts and documents and asks for any missing details.
Once the accountant confirms all the details, the AI assistant sends an email to the admin team with all the details nicely organised.
Do you see the pattern?
All these cases are around an AI Agent finding details in long, convoluted emails and documents, asking for missing details, confirming with the user, and neatly organising the details: in a template, a draft email or even sending an email.
Building and Sharing in ChatGPT
In ChatGPT, you create a new project.
The heart of the project is the instructions, the prompt.
It tells the project what role to play, what steps to take, and how to format the output.
The big news now is you can share these projects.
Custom GPTs were already shareable, but projects are now getting more attention.
They may well be the future of sharing prompts internally.
Microsoft Copilot: Notebooks, Agents and Studio
In Microsoft Copilot we have three options to achieve a similar result:
Option 1. We can use Notebooks, which work like ChatGPT projects with prompts and attachments.
We can’t share notebooks yet, but that’s likely coming.
Option 2: We can create lite agents in Copilot chat.
They work like notebooks but are shareable with your team.
Option 3: We can create advanced agents in Copilot Studio. Here we can go further by giving agents tools, like the ability to send emails and publish them in different locations in our Microsoft environment.
A Word of Caution Before You Share
When you publish a project or agent to your team, it's natural for the team to assume it’s reliable just because you tested it.
But it’s still AI.
It won’t always get things right, and everything still needs to be verified.
Make sure to communicate this clearly to your team.
Implementing at Your Firm: The AI Champions Club Process
Here’s the process I used in the last AI Champions Club session:
Step 1: Talk through the process. Clarify that the goal is to find details from documents, transcripts, or emails, spot what’s missing, and organise it into a usable output. Show examples. I showed my three examples above in great detail.
Step 2: Brainstorm use cases. Score them by time saved and risk level. Choose the best one for each team to experiment with.
Step 3: Write the prompt. Use a template that sets the role, lists the steps, and specifies the data, behaviour, and output format.
Step 4: Iterate to improve the prompt. When it is good enough, publish to your team to embed in workflow.
Join Us to Build Together
If you want to do this together with me, join the AI Champions Club.
We meet once a month and focus on AI capabilities, tools and use cases.
We support each other, share ideas, and put AI into workflow.
I'd love to have you there: inbal.com.au/community.
—-
Inbal Rodnay
Guiding Firms in Adopting AI and Automation
Keynote speaker | AI Workshops | Executive briefings | Consulting CIO
Want to receive these updates straight to your inbox? Click here: www.inbal.com.au/join
When you are ready, here is how Inbal can help:
Bring your entire team on the AI journey in just 30 days. This program is designed to give your team a solid foundation in using generative AI in responsible and impactful ways. Inbal helps you choose your AI tools, create an AI policy and train your team.
Want the confidence to set strategy and lead but don't have time to keep up with all the changes in tech?
Tailored for your needs, Inbal will works with you through one-on-one sessions to develop your technology literacy and keeps you up to date.
For CEOs, partners and business leaders. Everything you need to know about AI without the noise. Inbal shares the state of AI, recommends tools, and answers your questions about strategy, implementation and safe use.
Only what's real, no hype, no noise.
This is a one-off session for your entire leadership team.
