Problem statement: Organizations want AI and machine learning to understand the context of their business so that it can give answers from the organization’s vast data set. The bigger the corporations are, the more mature they tend to be with automation, analytics, and the more prepared they can be in terms of the data set. But, what about businesses who may not have the skillsets required to harness the AI and the data? Google Cloud’s technologies may be able to help with that.

A long line up of Google Cloud’s customers have hopped aboard when its end-to-end machine learning platform, Vertex AI, began to offer new generative AI capabilities. Brands like Fox Sports, Bank BRI, Toyota, Culture Amp, Estee Lauder, GE Appliances, Wendy’s and more, can take advantage of generative AI using the same platform and interface they are already familiar with from Google’s Cloud AI Platform.

It has become easier for customers to access foundation models and customize them with enterprise data to quickly build gen AI apps. As a result, the number of such projects on Vertex AI has grown over 150 times from April to July, this year, according to Google.

When talking with Caroline Yap, MD of Global AI Business at Google Cloud, it was evident that the promise of unleashing pent up creativity and innovation in a diverse range of industries can be infectious. 

“Banks can now decide if they want to use AI as the way to change the experience for customers,” she said after sharing an example of how a bank can use geolocation data and perceived customer needs (via machine learning) to inform customers via a mobile app, the nearest ATM which has the denomination they would need.

Frictionless quick service

“We all want to save time. How can we use AI to help give people time-savings from an end-customer experience perspective while also looking at employee productivity?” 

Another example Caroline gave from the fast-food industry involves AI-based vision technology to detect and alert staff of tables that need clearing and cleaning rather than rely on a maintenance schedule.

She then shared, “So, what the CEO of Wendy’s wanted was to save his people time from doing manual chores that could be done or automated by AI, so they could spend time doing more customer-facing tasks like customer service, or helping an elderly find a seat, and so on.”

An exciting development from their collaboration is Wendy’s FreshAI which reduces miscommunication due to ambient noise and customers customizing their orders at Wendy’s drive-thrus.

These are just a few examples of the innovative problem solving that generative AI has enabled for brands from various industries.

With so many fast-paced developments in the area of generative AI, where does a business start? Caroline suggested for companies to determine what matters to their businesses and to look at AI from three perspectives; growth, efficiencies, and the future.  

When working with customers, Caroline discovered that generative AI has actually unified the siloed teams of data scientists, developers, customer experience agents, IT, and more around a common business goal.  

A simple decision that makes an entire business flow faster could be due to these different teams working, experimenting and iterating together on a convenient user interface that delivers this frictionless flow.

“Those are the things that (Different ways of working) have emerged the most because of generative AI, and I don’t think it’s a hype cycle per se, because I feel people are having really meaningful conversations now across the different teams,” Caroline emphasized.

But, what about businesses who may not have the skillsets required to harness the AI and the data? Google Cloud’s technologies may be able to help with that.

Two other use cases that come up a lot as well for generative AI is frictionless interaction (conversations) between businesses, brand and end users, as well as easier search of complex documents and systems without one having to be a data scientist with querying skills.

Skills that organizations can start with

Caroline observed two data fitness levels that determine the skillsets required in organizations.  “It also depends on the industry. Some are already very mature with automation and analytics and have fairly decent data, with very mature machine learning models and more, that give them good returns.”

On the other end of the spectrum, are the organizations who don’t necessarily have a big team of data scientists or even machine learning experts. Tools and technologies like Google Cloud’s Duet AI address this by enabling database engineers for example, to perform (and learn) machine learning tasks like building a machine learning model by simply querying in natural language. 

Interestingly, Caroline observed from a skill sets perspective also, organizations are having to ensure that the different roles interacting with AI via search or chat, know how to do prompt engineering and prompt design.

Free courses via Google also offer a way for non-AI talent to upskill themselves. “So, it’s been really good to see everyone coming together to solve human-centric problems,” Caroline concluded.

SUMMARY

Different ways of working have emerged as a result of integrating machine learning and AI into the work flow.  The biggest benefit seems to be the breaking down of silos as different skillsets, experiences, and expertise start to communicate and blend with each other.