Yesterday, I led an all-day workshop for a consulting firm based in New York City focused on establishing a foundation around the core domains of AI, followed by hands-on training to apply generative AI to use cases for their client base.

Before getting started, there was a quick survey of familiarity with generative AI, as this domain was the majority focus for the day.

Participants' Use of Generative AI

In terms of how participants have used generative AI:

  • All have used it in some capacity for their work.
  • 75% have used it in their personal lives.

When asked to describe their sentiment in terms of a single word or short phrase, it was moderately positive on average, with no one having a negative perception.

Clients' Interest and Awareness

Their clients had a wider variation in reported interest and awareness:

  • Expressed interest in technology, but did not identify use cases.
  • Had interest but are risk-averse, desiring to tread slowly.
  • Unaware of what it is and how it could be applicable.

After discussing the handful of domains that are widely applicable to organizations (see here for a refresher), we focused on single domain and multiple domain examples.

Examples of Use Cases

The latter effectively demonstrates how a pipeline can exist. Here are two of the examples:

  • Asking a chatbot for recommended products given a set of inputs, then crafting messaging about why those are suitable.
  • Using computer vision to extract information from documents to pass into an algorithm developed with machine learning to predict whether it is a fraudulent claim.

Hands-On Portion

The last part of the educational material before going into the hands-on portion identified use cases for their clients’ industries and discussion about clients that could benefit from each use case, along with variations.

For the hands-on portion, participants built out three use cases leveraging generative AI and data from CRM:

  1. Defining an engagement strategy and identifying next best actions.
  2. Writing call talking points, meeting agendas, and drafting emails that align with the strategy or next best action.
  3. Generating hyper-personalized content to inject into email marketing campaigns.

All of the participants extended the core use cases to have their own take on how to combine data for context and prompts.

Core Takeaways

By the end of the day, we landed with these core takeaways:

  1. Every participant identified multiple clients that could benefit from one or more of the hands-on use cases around generative AI.
  2. Most participants identified at least one client that could use robotic process automation (RPA), even if used in a “semi-automatic” context.
  3. Most participants have worked with one or more clients that are plagued with manual data entry, and described use cases where computer vision (specifically extraction of data from handwritten/typewritten documents) would be a large efficiency boost.

TLDR: Look around to see how AI could improve your work and personal life, then share with your colleagues and ask them to do the same.