It’s so great to be back in person at San Francisco’s Moscone Center, the first time Google Cloud Next has been a large in-person event since 2019! We kicked things off with a bang, with a keynote from Google CEO Sundar Pichai, who talked about the work we’re doing with our AI technologies to make them more useful and responsible. Then, Google Cloud CEO Thomas Kurian laid out how customers are using our AI, infrastructure, and tools to rethink how they run their businesses and deliver new experiences to customers.
Those led into Spotlight and breakout sessions, the show floor, and many announcements.
Here are some highlights from Keynotes and Spotlight sessions that ran on day one:
AI everywhere
If you take away just one thing from day one of Google Cloud Next, it’s that we’re serious about AI. We showed off our always-on AI assistant Duet AI in Google Workspace and in Google Cloud. We updated our AI platforms with extensive new models and capabilities to help you create AI applications that are smart and manageable. And we optimized existing services and infrastructure to take advantage of AI — or help you train and serve it better.
What we announced:
Duet AI in Google Workspace is generally available, and Duet AI in Google Cloud is in preview. When it comes to the latter, you’ll be able to use Duet AI in Google Cloud services including BigQuery, Looker, Spanner, Database Migration Services, GKE, Mandiant Threat Intelligence, Chronicle Security Operations and Security Command Center.
Vertex AI is your one-stop shop for building generative AI apps, machine learning development, and MLOps. Updates includes new generative AI foundation models in Model Garden, including Meta’s Llama 2 and the pre-announcement of Anthropic’s Claude 2; updates to our first-party PaLM 2, Codey, and Imagen models; the launch of Colab Enterprise for your data-science workflows; and Vertex AI Extensions — tools to customize foundation models, combine them with your data, and build gen AI apps.
The new Cloud TPU v5e lets you train your AI workloads quickly and cost-effectively, and A3 VMs based on NVIDIA’s H100 GPUs are now GA.
We are the first cloud provider to enable digital watermarking and verification for images generated by AI, with technology based on Google DeepMind SynthID that helps users more easily identify when they’re encountering AI-generated content online.
Data analytics and database news galore
AI is only as good as the data that it’s trained on, and that data, by and large, lives in operational and transactional databases. To help, we unveiled a wealth of new capabilities and features to our core data and data analytics offerings.
What we announced:
The new BigQuery Studio presents a single interface for data engineering, analytics, and predictive analysis. Additional features provide data teams with a simplified lakehouse experience, cross-cloud analytics, secure data sharing, and improved governance.
AlloyDB now includes AlloyDB AI, which can help you build a wide range of AI experiences by making it easy for you to generate vector embeddings from within your database, access data stored in open-source gen AI tools such as LangChain, as well as local models in AlloyDB and remote models in Vertex AI.
Speaking of AlloyDB, we’re serious about helping you move your legacy databases to the PostgreSQL-compatible database service. Duet AI in Database Migration Service, now in preview, provides AI-assisted code conversion to automate the conversion of Oracle database code such as stored procedures, functions, triggers, packages and custom PL/SQL code.
Infrastructure and security FTW
Underpinning all Google Cloud capabilities — AI and otherwise — are core infrastructure and security offerings, which we are constantly building on and improving.
What we announced:
We updated our Google Distributed Cloud portfolio, including GDC Hosted and GDC Edge, to support a bevy of new services on new-and-improved hardware.
The new Cross-Cloud Network global networking platform includes products from Google Cloud’s networking portfolio as well as from partners’ to help simplify connectivity between clouds, enhance security, and improve networking performance.
Mandiant Hunt for Chronicle provides managed threat hunting by Mandiant experts on Chronicle data to help expose attacker activity and reduce business impact.
Security Command Center’s agentless vulnerability scanning, powered by Tenable, can detect operating system, software, and network vulnerabilities on Google Compute Engine VMs.
Not to be missed: network security advancements in Google Cloud Firewall, more processor support for Confidential Computing, Japan Regions support for Assured Workloads, and more.
A Google Workspace wonderland
In Workspace world, the headline news was that Duet AI in Workspace is generally available, but other elements of the Workspace family got plenty of love, too.
What we announced:
Leveraging Google AI, we enhanced Workspace with several new zero trust and digital sovereignty controls. Read about all the Workspace security announcements we made here.
It’s not your grandfather’s Google Chat: New innovations in Google Chat, including Duet AI in Chat, can deliver insights from documents shared in a space, provide summaries of conversations, and more. Read all about it here.
Better together
Finally, where would we be without you — our customers and partners? To celebrate, we announced the winners of our annual Google Cloud Customer and Partner of the Year awards, delivered new partner services, and unveiled a long list of AI startups building on our platform.
And that’s just the beginning! Today marked 18,000 people coming together for the first of three great days of keynotes, breakouts, trainings, networking, and hobnobbing. See you again tomorrow, when we’ll recap the news coming from our day two keynotes and sessions, including the ever-popular developer keynote. And if you weren’t able to make it to the show in person, be sure to register for a complimentary Digital Pass and watch keynotes and breakouts on demand!
Cloud BlogRead More
