Making Artificial Intelligence Helpful for Everyone
Google Cloud goes all in on generative AI at Next ‘23
If you were hoping to get Google Cloud’s take on generative artificial intelligence (generative AI, or GenAI) during its recent Google Cloud Next ’23 conference in San Francisco, you certainly didn’t come away disappointed. Of all the sessions now free to view on the event’s website, over half include some form of GenAI content. But even for those of us who knew that GenAI would be hogging the spotlight, there were still some surprises and reasons to be excited about how Google Cloud is approaching what’s become the number one topic in tech circles and beyond.
For an event of this magnitude, where one might expect to be overloaded with new product, feature, and functionality information, it was actually incredibly easy to home in on the key messages Google Cloud wanted its customers to come away with:
- Generative AI needs to be made accessible to everyone.
- Google Cloud is committed to being responsible when it comes to GenAI technology.
- With its world-class infrastructure and ecosystem, Google Cloud is uniquely positioned to help its customers run generative AI programs safely and efficiently.
Let’s take a look at each of these in a little more detail.
All for GenAI, and GenAI for all
Google Cloud wants to make GenAI available to everyone, and it’s putting its money where its mouth is by providing the types of innovations to make this possible.
To start with, Duet AI, which was officially announced in May, will be generally available in Google Cloud later this year. But in its current preview version it already boasts capabilities to help with application development, operations, database management and migration, data analysis and visualization, as well as cybersecurity. Some of the incredible applications for this technology include:
- Converting older code languages to newer ones for application modernization and migration. More and more organizations are moving away from outdated infrastructures to more modern, user-friendly environments. Using generative AI to do a lot of the heavy lifting involved in this process is a game changer. Slalom’s expertise in this area has earned us a Google Cloud Migration Specialization and made us a trusted partner in Google Cloud’s rapid migration program, or RaMP.
- Generating context-aware code using natural language prompts directly in your development environment. Developers can use their organizations’ own code repositories as examples when asking AI to draft new code.
- Speeding data analysis within Looker and BigQuery to easily create calculations and visualizations without needing to write any code. Brad Calder, VP and GM, Google Cloud Platform and Technical Infrastructure, writes, “With Duet AI in Looker we enable conversational data analysis, automatic generation of presentations, intelligent text summaries based on your reports, the use of natural language to create calculations and visualizations, as well as the ability to rapidly generate LookML models.”
But even if you’re not a developer or data scientist, you can still reap the benefits of Duet AI. Google Workspace, which includes productivity and collaboration tools such as Google Docs, Sheets, Slides, Meet, and, of course, Gmail, now also comes with the generative AI capabilities of Duet AI.
In the opening keynote, Aparna Pappu, GM and VP of Workspace, explained how this could work in Google Meet. If you join a meeting late, Duet AI has an interactive “catch me up” mode that provides a summary of everything you’ve missed. You can even ask follow-up questions on specific topics that were discussed. Duet AI can also detect and translate over 300 language pairs and provide real-time captions for meeting participants.
This is just one example of the incredible power of this technology. Others include creating text and image content, delivering insights from data in Google Sheets, auto-drafting Google Slides decks based on information in Google Docs and applying your organization’s branding and communication standards to those slides … The list goes on and on. And by extending generative AI functionality to tools used across the organization — not just by your IT and data teams — Google is taking an important step toward fulfilling its promise of making AI helpful for everyone.
Responsible GenAI
No Google Cloud Next conference would be complete without some “ooh” and “aah” moments, and this year’s event certainly was no exception. Many of these moments occurred during the conversation about responsible AI.
Vertex AI is a tool used to train and deploy AI models and applications. A new feature within the tool has been designed to help users more easily identify when they are encountering AI-generated content. Using technology powered by Google DeepMind SynthID, images generated by Vertex can be watermarked in a new and safer way. Traditional watermarking has either been done within the image’s metadata or with a visible pixel-based stamp that could be clearly seen on the image. Both of these methods have proven easy to circumvent. But the digital watermarking method now in preview in Vertex AI is more secure while at the same time invisible to the human eye.
Verification tools can be used to search for these watermarks and identify whether the image was generated by AI. Google Cloud is the first cloud provider to enable digital watermarking and verification for images generated on its platform.
During the “What’s new with generative AI” spotlight session, June Yang, VP, Cloud AI and Industry Solutions, Google Cloud, outlined some of the other ways Vertex AI is helping customers achieve more responsible GenAI outputs. A new grounding feature works across Vertex AI custom models, search, and conversations by connecting with your enterprise data to deliver more accurate responses and help reduce model hallucinations. It forces the model to create output that’s based on user-specified source material that’s considered to be more factual and verifiable than sources used in many large language models. Vertex AI Search (previously known as Enterprise Search on Generative AI App Builder) is now also grounded in your enterprise data with auditable citations, which is huge from a trust and verification perspective.
Bringing substance to the flash
While the demos of the new features and capabilities are certainly a big part of any major tech conference like Google Cloud Next, the event’s speakers also gave plenty of thoughtful attention to the underlying structures that enable those amazing achievements. These announcements fell under the categories of security, infrastructure for AI workloads, databases, networking, and many others. Amid these more technical elements — which are vital to the success of your generative AI program — were also some more human elements.
In the keynote, Thomas Kurian, CEO of Google Cloud, laid out Google Cloud’s three-pronged approach for a robust generative AI program, which includes Vertex AI for building AI-enabled technologies, Duet AI for using AI assistants within Google Cloud’s business applications, and also a strong ecosystem “to ensure that every organization can succeed in adopting AI.” Google Cloud’s work with its ecosystem of partners, customers, and startups is vital to this success.
In an on-site conversation with enterprise tech experts from theCUBE, Slalom’s Tony Ko (managing director, Global Data, Analytics and AI) and Cameron Curtis (senior director, Global Google Center of Excellence) note that because GenAI technology is changing and evolving at a pace that we’ve never seen before, working with Google’s partners is more important than ever. Slalom’s Google Generative AI Center of Excellence was formed to further nurture an ecosystem of experts to find new ways to leverage Google Cloud’s AI technologies effectively for our customers.
Closing the gap
At Slalom, we recognize that when it comes to running a successful GenAI program, many organizations face a wide gap between expectations and understanding. As part of Google Cloud’s partner ecosystem, we can help you close that gap through discovering how you can use generative AI to:
- Boost productivity and the efficiency of many of your organization’s processes,
- Analyze the entire breadth of issues affecting your business to come up with prescriptive recommendations for solutions, and
- Further flex your critical thinking muscles to challenge any remaining constraints and achieve radical transformation of your business.
Learn more about Slalom’s partnership with Google and how we’re bringing generative AI to life for our customers.
Slalom is a global consulting firm that helps people and organizations dream bigger, move faster, and build better tomorrows for all. Learn more and reach out today.
