Context is Everything in Generative AI

There is a lot of chatter about whether companies are realizing value from generative AI. Realizing value from generative AI will take time to prepare, capture, and create the data that allows generative AI to understand the context of our complex human lives.

An abstract image of a large pile of docum

For decades, companies have decontextualized complex information to create usable and transferable data. Customer experiences are captured as metrified moments that remove the context of the human experience to be most useful when making recommendations, conducting transactions, or servicing complaints. Generative AI, however, opens a new world of opportunity to use unstructured, contextualized data, exposing that data through conversational interfaces. The challenge is: how to prepare or create this data for generative AI.

We think about this challenge in three phases:

  1. Prepare. Currently, the rush is to prepare existing data for generative AI. While 93% of Chief Data Officers and data leaders agree that data strategy is critical for getting value from generative AI, only 11% agreed strongly that their organizations have the right data foundation for generative AI. That gap highlights the amount of work to be done to prepare and expose current data to generative AI models. There are many challenges in this process but the one that stands out the most for us is governance. Unstructured data provides highly valuable context, but that context may also be sensitive or confidential. Mapping who can have access to this information—internally and externally—creates a new and novel governance challenge.
  2. Capture. Once an organization has figured out how to prepare and use its current data, the next question will likely be: how can we capture more contextualized data to improve our generative AI? While the governance processes hopefully translate to new data capture, data leaders will need to create new, novel methods, workflows, and funnels for capturing data that likely don’t exist today.
  3. Create. The final and most lucrative step will likely be in finding new ways for creating data that doesn’t exist today. This will require reconsidering and redesigning customer experiences. Conversational interfaces will break current product and experience barriers, meeting customers where they are in their journey. This is a profound change that can be both a threat and an opportunity. For instance, the travel industry may experience a dramatic shift as Google creates opportunities for users to organize itineraries and book travel from within the generative AI search experience. Google is best positioned to understand the context of a user’s travel interests by shifting search prompts to search conversations. And these conversations will create new contextual data that doesn’t exist today.

There is a lot of chatter about whether companies are realizing value from generative AI. As we discussed last week, some of that is due to the nature of complex change. An additional aspect, however, is that realizing value from generative AI will take time to prepare, capture, and create the data that allows generative AI to understand the context of our complex human lives.

We’ll be talking about context, complexity, and generative AI at the EY Innovation Realized conference on April 17th in San Francisco. Reach out if you’ll be at the conference or nearby in San Francisco—we’d love to connect in person.

We'll also be sharing our latest summary of generative AI progress across all fronts with a particular focus on how productivity and adoptiong is going for laggards and leaders on April 9th in our State of AI and Complex Change Report: Q2 2024 Update. This event is open for all subscribers: keep a watch out for a separate email with Zoom login info.

Interested in talking more about how we might help your organization navigate the new worlds of AI and complex change? Set up time for a chat with us here.

This Week from Artificiality

  • A Leader's Guide to Navigating the Generative AI Revolution. We present a guide for organizational transformation for generative AI in five steps. 1) Recognize that Generative AI Adoption is Cultural, 2) Adapt to Complex Change from Generative AI, 3) Develop "Minimum Viable Adoption" Guidelines for AI, 4) Decide What Needs to be Centralized, and 5) Train, Train, Train. By taking these critical steps, organizations can lay the foundation for effective use, setting themselves up for future success in an increasingly AI-driven world.
  • AI and the Extended Mind. Continuing our series on a Mind for our Minds, we ask: Does it matter where our thinking is being done? How should we as individuals think about outsourcing more of our cognition to AI? We believe the future of the extended mind is not a fixed destination but an ongoing journey that requires our active participation and reflection as we redefine what it means to be human in an age of artificial intelligence.
  • Conversing with AI, Part 4: Segment. In the next installment of our toolkit for conversing with AI, we show how to use generative AI to segment a problem, aka modularizing tasks with generative AI to make them more achievable. If you haven't checked out our generative AI toolkit before, rewind to Part 1, Part 2, and Part 3. And stay tuned for more tools for your toolkit coming soon.

Bits & Bytes from Elsewhere

  • AI anxiety is solidly in the public consciousness and our favorite example is this week's segment on The False Promises of AI from Jon Stewart on The Daily Show. Take 15 minutes to watch Jon's take on AI misinformation and his humorous critiques of self-described "utopic tech bro"s like Marc Andreessen, Sam Altman, and Mark Zuckerberg. While they talk about the promise of AI curing diseases and solving climate change, Jon makes fun of Zuck showing off that his AI, Jarvis, can make toast. Jon shows clips of other tech leaders professing that AI will be an assistant for everyone but then also saying that we will experience "overall displacement in the labor market," that AI will enable the "same work done with fewer people," that AI will provide "productivity without the people tax," and that AI products are "fundamentally labor-replacing tools." Jon finishes by highlighting that if a prompt engineer (a role he says is actually just a "types-question guy") is the job of the future, then our future will be about people working for AI, rather than the other way around.
  • A customer data security risk was found at Hugging Face by cloud security company, Wiz. While the risk was fixed before any damage was created, this highlights the risk of a) working with new, startup vendors and b) exposing new data sources through generative AI. As we talked about above, unstructured data is essential to realizing the true benefits of generative AI. But that unstructured data can also include context, comments, and perspectives that would be especially damaging if released into the open (just consider the damage of emails released through trials at various big tech companies).
  • Perplexity plans to sell ads in its generative search product. The company previously said that search should be "free from the influence of advertising-driven models." Perhaps its finding that there aren't enough consumers willing to add another gen AI subscription. Perplexity's current model is more about providing answers itself than providing links to answers which raises questions about how ads will be presented and what advertisers will pay for. Will the ad be embedded in text generated by Perplexity? Will advertisers find value without users clicking on a link to their own sites?

Where in the World are Helen & Dave?

Several upcoming events to highlight—you can see everything on our events page. 

  • ArtificialityState of AI and Complex Change Report: Q2 2024 Update. Join us on April 9th for our new, quarterly research update on AI and complex change. Reach out with any questions. 
  • Charter ProMaking Decisions with Generative AI. Join us on April 10th for a presentation with our friends from Charter about making decisions with generative AI. Sign up here.
  • ASU+GSV Summit. Join us on April 15th at the ASU+GSV Summit in San Diego. No presentations scheduled yet but we may do a pop-up or impromptu gathering depending for the Artificiality community. 
  • EY Innovation Realized. Join us on April 16-17 at EY Innovation Realized at the Palace of Fine Arts in San Francisco for our presentation on Rethinking Value in an Age of AI.
  • Starbucks Innovation Expo. Join us on May 14-15 as we return to the Starbucks Innovation Expo in Seattle for the fourth time to talk about Generative AI & Data Culture.

Great! You’ve successfully signed up.

Welcome back! You've successfully signed in.

You've successfully subscribed to Artificiality.

Success! Check your email for magic link to sign-in.

Success! Your billing info has been updated.

Your billing was not updated.