AI Adoption Advice from the Vercel Ship 2024 Enterprise Panel

At Vercel Ship 2024, an annual end-user conference, a panel of AI experts from Google, Groq, and OpenAI convened to discuss how enterprise teams are deploying AI technologies. The conversation, moderated by Sabrina Halper of Tomorrow Talk, covered the range of outcomes businesses are pursuing: automating tasks to boost productivity, building personalized user experiences, and exploring business models that weren't feasible before.

While the specifics of each customer deployment varied, the panelists converged on a set of foundational principles for teams that are just beginning their AI journey.

Reimagine Processes Before Automating

The first major theme was the need to think beyond digitizing existing workflows. AI's real value, the panel argued, lies in its capacity to enable entirely new interactions. For instance, the web could move beyond the confines of a two-dimensional screen into ambient, three-dimensional spaces that persist around us throughout the day, adapting to our context and needs.

"There's so much potential here. And it's not just for having the web be confined to this two dimensional screen that we carry around with us, but actually being able to have it persist in the world in three dimensional space, and being able to be ambient with us as we experience our days."

Paige Bailey Google

Focus on User Pain Points, Not Automation

A related recommendation was to let user needs serve as the north star. Instead of seeking out tasks to automate for the sake of efficiency, teams should identify real pain points that are amenable to scaling with AI. The panel pointed to healthcare as a powerful example, citing a case where AI helped a patient regain their voice after losing it to illness; a direct response to a profound human need.

Fix Your Data First

Before evaluating specific AI models or tools, the experts issued a clear warning: get your data in order. Investments in AI will be undermined by unstable or poorly managed underlying data. Establishing rigorous data quality practices and management processes is the critical groundwork that ensures accuracy and long-term success.

"We always tell people, start with your data. This is the chance for you to fix your data models or even your data management. Do that work first because then it'll really pay off."

Sunny Madra Groq

Build an Evaluation Framework

Another key piece of advice was to establish a strong internal framework for evaluating AI initiatives. This means raising awareness across the company and empowering subject matter experts to identify promising use cases. A structured approach allows enterprises to compare models, track progress over time, and pinpoint areas for improvement, helping ensure that AI deliverables on its promises while reducing undesired outputs.

Prioritize Transparency and Human Oversight

Finally, the panel stressed the importance of transparency for building user trust. People are more likely to accept AI-driven insights when they can understand the reasoning behind them. Keeping a human in the loop, with AI offering options and recommendations rather than making autonomous decisions, is a recommended safeguard.

"The ingenuity doesn't stop, and teams are really pushing the boundaries."

Miqdad Jaffer OpenAI

The overarching message was that AI has immense potential to transform enterprises, but success depends on a deliberate, human-centered approach. By reimagining processes, prioritizing user needs, establishing strong data and evaluation practices, and embracing transparency, teams can lay the groundwork for a future where AI is seamlessly woven into our daily lives.