This week in Las Vegas, 30,000 people gathered to hear the latest and greatest from Google Cloud. What they heard was always generative AI. Google Cloud is first and foremost a cloud infrastructure and platform vendor. If you didn't know that, you might have missed it amidst the onslaught of AI news.
Not to discount what Google had on display, but like Salesforce at its New York City roadshow last year, the company couldn't do anything other than nods to its core business — not to mention generative AI. (out of context).
Google announced a number of AI enhancements designed to help customers leverage Gemini large-scale language models (LLMs) and improve productivity across the platform. Of course, this is a worthy goal, and throughout the main keynote on the first day and the developer keynote the next day, Google peppered the announcement with a number of demos to illustrate the power of these solutions.
However, many seemed a little too simple, even considering the limited time that needed to be crammed into the keynote. They primarily relied on examples within the Google ecosystem, as nearly all companies store much of their data in repositories outside of Google.
Some examples actually felt like they could have been done without the AI. For example, during an e-commerce demo, the presenter called a vendor to complete an online transaction. This was designed to show off the sales bot's communication capabilities, but in reality, this step could have been easily completed by the buyer on her website.
That's not to say that generative AI doesn't have powerful use cases, such as being able to write code, analyze and query corpora of content, and interrogate log data to understand why a website is down. Additionally, the task- and role-based agents the company has introduced to assist individual developers, creative people, employees, and more have the potential to leverage generative AI in tangible ways.
But when it comes to building AI tools based on Google's models, rather than leveraging the AI tools that Google and other vendors are building for their customers, they face a number of potential hurdles. I couldn't help but feel like I was being ignored. How to successfully implement generative AI. Although they tried to make it sound easy, the reality is that implementing advanced technology within large organizations is a huge challenge.
Big changes aren't easy
As with other technology leaps over the past 15 years, innovations such as mobile, cloud, containerization, and marketing automation have come to fruition and promised many potential benefits. But each of these advances brings its own level of complexity, and large companies are moving more carefully than we might imagine. I feel like AI is making much more progress than what Google and frankly other big vendors are doing.
What we have learned from past technological changes is that they come with a lot of hype and a lot of disillusionment. Years later, large companies that should be taking advantage of these advanced technologies are still doing little or not using them at all, even years after they were introduced. we have seen.
There are many reasons why companies fail to take advantage of technological innovations, including organizational inertia. A weak technology stack that makes it difficult to adopt new solutions. Or a group of corporate naysayers who shut down even the most well-intentioned initiatives, whether it's legal, HR, IT, or any other group that just keeps saying no to substantive change for a variety of reasons including internal politics.
Vineet Jain, CEO of Egnyte, a company focused on storage, governance, and security, sees two types of companies. One is companies that have already made significant migrations to the cloud, and the other is companies that will be able to easily adopt generative AI. And there are some companies that have been slow to move and are likely to struggle.
He speaks to many companies that still run large parts of their technology on-premises and have a long way to go before they can start thinking about how AI can help them. “We talk to a lot of 'late' cloud adopters who have not yet begun their digital transformation journey or are in the very early stages,” Jain told TechCrunch.
The advent of AI may force these companies to think seriously about undertaking digital transformation, but they may struggle from a very slow start, he said. “These companies need to solve those problems first and then leverage AI once they have mature data security and governance models,” he said.
it was always data
Big vendors like Google make it sound like these solutions are easy to implement, but like any advanced technology, what looks simple on the front end isn't necessarily complicated to implement on the back end. Not necessarily. As we've heard a lot this week, it's still a “garbage in, garbage out” situation when it comes to the data used to train Gemini and other large-scale language models, and that's not the case when it comes to generative AI. Even more applicable.
It starts with data. Without a data house in place, it becomes very difficult to get an LLM into a shape that can be trained for your use case. Kashif Rahmatullah, a principal at Deloitte who runs his Google Cloud practice at the firm, said he was primarily impressed by his Google announcements this week, but some companies lacking clean data It was acknowledged that there are problems with the implementation of generative AI solutions. “Sometimes these conversations start with AI conversations, but they quickly turn into: ‘We need to fix the data, we need to clean the data, we need to put all the data in one place, or almost You need to get it all in one place and then modify the data to start reaping the real benefits from generative AI,” Rahmatullah said.
From Google's perspective, the company has built generative AI tools that more easily help data engineers build data pipelines to connect to data sources inside and outside the Google ecosystem. “This is aimed at speeding up data engineering teams by automating many of the highly labor-intensive tasks involved in moving data and serving these models,” he said. said Gerrit Kazmaier, Vice President and General Manager, Looker. At Google, he told TechCrunch.
This should help connect and clean data, especially for companies furthering their digital transformation efforts. But for companies like the ones Jain mentioned, those that haven't taken meaningful steps toward digital transformation, even these tools created by Google can create additional challenges.
All of this even takes into consideration that AI comes with its own set of challenges beyond pure implementation, whether it's an app based on an existing model or specifically if you're looking to build a custom model. No, says analyst Andy Thurai. Constellation research. “When implementing any solution, companies need to think about governance, liability, security, privacy, ethical and responsible use of their deployment, and compliance,” Turai said. And none of it is easy.
Executives, IT professionals, developers, and others who attended GCN this week may have gone looking for what Google Cloud has to offer next. But if they didn't go looking for AI, or they just weren't ready as an organization, they might have left Sin City a little shocked that Google was going all-in on AI. yeah. It may take a long time for organizations that lack digital sophistication to be able to take full advantage of these technologies beyond the more packaged solutions offered by Google and other vendors. There is a gender.