Google works on generative AI on Google Cloud Next

this week in las Vegas, 30,000 people came together to hear the latest and greatest from Google Cloud. All they heard all the time was generative AI. Google Cloud is first and foremost a cloud infrastructure and platform vendor. If you don’t know this, you may have missed it in the onslaught of AI news.

Not to diminish what Google displayed, but like Salesforce at its New York City travel roadshow last year, the company failed to acknowledge its core business — except in the context of generative AI, of course.

Google announced several AI enhancements to help customers take advantage of the Gemini Large Language Model (LLM) and improve productivity across the platform. Undoubtedly, this is a worthy goal, and during the keynote on the first day and the developer keynote the next day, Google included announcements with a large number of demos to illustrate the power of these solutions.

But many seemed a little too simplistic, even taking into account what they needed to incorporate into the keynote speech with limited time. They relied mostly on examples from inside the Google ecosystem, when most of almost every company’s data was in repositories outside Google.

Some examples really felt like they could have been done without AI. For example, during an e-commerce demo, the presenter called on the seller to complete an online transaction. It was designed to show off the communication capabilities of the sales bot, but in reality, this step could have easily been completed by the buyer on the website.

This doesn’t mean that generative AI doesn’t have some powerful use cases, whether it’s creating code, analyzing a collection of content and being able to query it, or being able to ask questions of log data to understand what’s going on. Why did a website shut down? Additionally, the task and role-based agents the company has introduced to help individual developers, creative people, employees, and others have the potential to leverage generative AI in concrete ways.

But when it comes to building AI tools based on Google’s model, as opposed to consuming the tools Google and other vendors are building for their customers, I couldn’t help but feel that they were facing a number of hurdles. Ignoring what may come in it. The path to a successful generic AI implementation. Although they tried to make it look easy, in reality, implementing any advanced technology inside large organizations is a big challenge.

big change is not easy

Like other technological leaps over the past 15 years – whether mobile, cloud, containerization, marketing automation, you name it – it has been introduced with lots of promises of potential benefits. Yet these advances introduce their own levels of complexity, and big companies proceed more cautiously than we might imagine. It feels like AI is a much bigger achievement than what Google or, frankly, any major vendor is offering.

What we’ve learned from these past technology changes is that they come with a lot of hype and cause a lot of disillusionment. Even after many years, we see big companies that should probably be taking advantage of these advanced technologies, but even years after they were introduced they are still just scrambling or sitting out altogether. Have happened.

There are a number of reasons why companies may fail to take advantage of technological innovation, including organizational inertia; a brittle technology stack that makes it difficult to adopt new solutions; Or a group of corporate adversaries who are shutting down even the best-intentioned initiatives, be it legal, HR, IT or other groups who, for a variety of reasons, including internal politics, continue to say no to substantive change.

Vineet Jain, CEO of Egnite, a company focused on storage, governance and security, sees two types of companies: those that have already made a significant shift to the cloud and those that have an easier time when it comes to adopting generic AI. There will be time. And those that are are slow moving and will likely struggle.

He talks to a lot of companies that still have most of their technology on-premises and have a long way to go before they start thinking about how AI can help them. “We talk to many ‘late’ cloud adopters who haven’t started exploring digital transformation or are doing it very early,” Jain told TechCrunch.

He said that AI can force these companies to think about digital transformation, but they may have to struggle by starting too far behind. “These companies have to solve those problems first and then consume AI while having a mature data security and governance model in place,” he said.

it was always data

Big vendors like Google make these solutions simple to implement, but like all sophisticated technologies, looking simple from the front doesn’t mean it’s simple from the back. As I’ve heard often this week, when it comes to the data used to train Gemini and other large language models, it’s still a case of “garbage in, garbage out” and when it comes to generic AI This applies even more when it comes to.

It starts with data. If you don’t have your data house organized, it will be very difficult to get it into shape to train an LLM on your use case. Deloitte Principal Kashif Rahmatullah, who is in charge of the Google Cloud practice at his firm, was impressed by Google’s announcements this week, but still acknowledged that some companies that lack clean data will need to implement generative AI solutions. There will be a problem. “These conversations may start out as AI conversations, but it soon turns into: ‘I need to fix my data, and I need to clean it up, and I need it all in one place, or almost one. need to be put in place before I can start getting real benefits from generative AI,” Rahmatullah said.

From Google’s perspective, the company has built generative AI tools to help data engineers more easily build data pipelines to connect data sources inside and outside the Google ecosystem. “It’s really meant to speed up data engineering teams by automating a lot of the labor-intensive tasks involved in moving data and preparing it for these models,” said Gerrit Kazmaier, vice president and general manager of databases, data analytics and Looker. ” at Google, told TechCrunch.

This should be helpful in connecting and cleaning data, especially in companies that are on a digital transformation journey. But for the companies Jain mentioned — those that haven’t taken meaningful steps toward digital transformation — this could present more difficulties, even with these tools made by Google.

All this doesn’t even take into account that AI comes with its own set of challenges beyond pure implementation, whether it’s an app based on an existing model, or especially when trying to build custom models, says Andy Thurai, a Analyst says Constellation Research. “When implementing any solution, companies need to think about governance, liability, security, privacy, ethical and responsible use and compliance with such implementation,” Thurai said. And none of this is trivial.

Executives, IT professionals, developers and others who visited GCN this week likely went looking for what’s next from Google Cloud. But if they didn’t go looking for AI, or they’re not prepared as an organization, they may have come away from Sin City a little surprised by Google’s complete concentration on AI. Beyond the more packaged solutions offered by Google and other vendors, it may take a long time for organizations lacking digital sophistication to take full advantage of these technologies.