Powering the next generation of AI startups with Google Cloud

Today, many of the most exciting innovations in generative AI are coming from fast-growing startups.  We’re proud that the majority of these companies are building on Google Cloud, using our open and optimized stack for AI; broad choice of models and compute infrastructure; and many important resources and routes-to-market tailored for startups.

In fact more than 60 percent of all funded generative AI startups, and nearly 90% of today’s generative AI “unicorns” (those valued over $1 billion) are Google Cloud customers.

At Google Cloud Next, we’re announcing many new and expanding startup customers who are using Google Cloud’s AI-optimized infrastructure, models, AI and ML tooling, and routes-to-market like Google Cloud Marketplace. These include companies who are building the tooling, platforms, and models that will be foundational to the future of generative AI, like Assembly AI, Gretel, Higgsfield, OctoAI, and Writer. We’re also showcasing new businesses being built with AI from the ground up, like Ohalo Genetics, and vertically aligned startups who are finding exciting ways to solve industry-specific issues with AI, like Atomo Health, Rad AI, and Queenly. 

In addition, we’re rolling out new initiatives and resources, including an exciting new partnership with NVIDIA, that will make it easier than ever for early-stage AI startups to get up and running on Google Cloud, and new ways for early-stage companies to get involved with the Google for Startups Cloud Program. Read more below for all the details.

Google Cloud is the cloud of choice for startups building with AI

Innovation from the world’s most exciting startup businesses will be on display at Google Cloud Next ‘24, including at our Startup Theater; in sessions featuring the founders of companies like Anthropic, AI21 Labs, Glean, and Hugging Face; and through hundreds of customer and partner demos across the event. We’re also thrilled to announce a group of new and expanding startup customers that are charting the future of AI on Google Cloud, including:

• Arize AI is launching a new capability — Prompt Variable Monitoring — designed to help AI engineering teams automatically detect bugs in prompt variables and surface problematic datasets when troubleshooting LLM-powered apps. This capability was built using Google Cloud and supports Vertex AI and Gemini models. 
• AssemblyAI is building new AI systems that can understand human speech with superhuman abilities. AssemblyAI uses TPUs on Google Cloud to lower the cost of inference at scale for the thousands of organizations building cutting-edge AI features in its Speech AI models. AssemblyAI's newest model, Universal-1, advances the state-of-the-art in multilingual Speech AI accuracy, and uses Google Cloud infrastructure and TPUs for both training and inference. 
• Atomo, Inc. is launching AskMED.ai™, a new generative AI platform engineered to deliver fast healthcare insights from real world data, using Google Cloud's BigQuery and Vertex AI for unparalleled precision and speed. 
• Augmented Intelligence (AUI), is developing its own foundation models, Controllable Language Models™, based on a new AI architecture that orchestrates small language models to achieve control. AUI’s new AI conversational commerce solution, which retailers and consumer companies can embed into their websites, offers a convenient, natural-language interface for consumers from search to checkout. In addition, AUI’s solution, Apollo, is now available on Google Cloud Marketplace, and its AI models will use Google Cloud compute, GPUs, and our underlying data infrastructure for AI.
• Connected-Stories, an end-to-end creative management platform, is launching a new feature, powered by Gemini models on Google Cloud, that allows users to generate dynamically personalized ads in seconds using only natural language.
• CrateDB is launching its managed service, CrateDB Cloud, on Google Cloud and is using Vertex AI and Gemini models to power advanced forecasting, anomaly detection, and gen AI capabilities in its data platform.
• Fulfilld has built its intelligent warehouse management application on Google Cloud, using Vertex AI. Targeted to both midmarket and large enterprise customers, Fulfilld’s platform helps optimize warehouse inventory management, product placement, and employee efficiency.
• Gretel is focused on helping developers generate high-quality synthetic data to develop and train AI models. Gretel offers its platform on Google Cloud Marketplace and now, Gretel is launching a new integration with BigQuery that will enable businesses to quickly access its synthetic data capabilities from within Google Cloud.
• Hiber has developed a lightweight gaming engine, Hiber3D, that allows game developers to build 3D worlds across platforms. Hiber is launching a new feature, called Hiber3D SkyScape AI, that gives developers the ability to create a 3D world simply by uploading a picture or photo, powered by Gemini models through Vertex AI.
• Higgsfield, a startup working on personalized AI video creation and editing capabilities, is now using Google Cloud to train its foundational video model, and has built a highly scalable, complex inference pipeline for multiple models on Google Cloud. Training requires transmitting a lot of data over the network, so Google Cloud provides reliable and robust solutions which, in turn, enable Higgsfield to continuously deliver new ways for its users to create richer and more personalized content.
• Moonhub is launching an AI-native recruiter to help startups hire and scale their teams more efficiently. Moonhub's foundational recruiter AI, built on Google Cloud, uses machine learning and natural language processing to help uncover hidden talent.
• Mozart Data is launching a major update to its data platform, built on BigQuery, that will help customers better prepare their data for analysis, visualization, and AI.
• OctoAI is announcing a strategic partnership with Google Cloud to bring its generative AI developer stack to our trusted, AI-optimized infrastructure, enabling greater scale and performance for OctoAI’s platform and providing even more capabilities to developers working on AI applications.
• Orby AI is bringing together enterprise automation and generative AI to deliver a new AI-powered automation platform that helps users continuously find new ways to streamline and automate common, repetitive tasks. Its AI automation platform is built entirely on Google Cloud. 
• Ohalo Genetics is a newly-announced startup focused on applying AI to agriculture. As the company grows, it has selected Google Cloud as its platform and AI provider of choice, helping it develop and scale important technologies for the agriculture industry. 
• PharmaGuide is launching a new chatbot, powered by Google Cloud AI and integrated into its PHOX platform, used widely by pharmacists to streamline common processes and provide efficient care to patients. Its new chatbot provides instant access to information and guidance to help surface timely, helpful information for pharmacists. 
• Queenly, a Y Combinator company that offers marketplace and search functions, is powering a new AI-generated virtual try-on experience with Google Cloud’s AI and data cloud capabilities. 
• Rad AI is applying AI models, trained on Google Cloud using GKE, to make important advances in its lung cancer screening mode through its Continuity platform.
• Reality Defender builds software that helps in the detection of deepfakes and AI-generated disinformation. It is now launching a real-time voice deepfake detection platform, built on Google Cloud and using NVIDIA A100 hardware, that will be used by financial services and call centers. 
• Rocket Doctor, the digital health platform and marketplace, is rolling out new AI features built with a suite of products from Google Cloud, including Vertex AI, MedLM, and data analytics tools, which will help doctors intelligently search and summarize patient data in their EHR systems. 
• Silvr is a France-based tech startup that provides a platform for financing for small- and medium-sized businesses. Silvr is now using Google Cloud’s AI capabilities to speed up the financing process by automating tasks like credit capacity analysis and real-time notation capabilities.
• Snorkel AI is launching Snorkel Custom, an offering that combines Snorkel’s programmatic AI data development platform — Snorkel Flow — with hands-on support from Snorkel’s machine learning experts to help enterprises use their data to adapt LLMs and deliver production quality AI faster. Snorkel also expanded native integrations for Google Cloud LLMs, announcing new support for Gemini models.
• Suggestic, a turnkey platform for launching tele-wellness applications, is now launching a new product, Viium AI, that delivers highly personalized GLP-1 weight loss experiences, improves patient outcomes, and boosts provider revenue. Viium AI is built using GKE, Vertex AI, and Gemini models. 
• Swit is using Google Cloud’s AI models through Vertex AI to power a new Snap chatbot product that can help people simplify common tasks at work, like managing to-dos, building checklists, creating contextual responses, and more.
• VEED.IO is launching a new AI text-to-video tool that turns simple text prompts into engaging videos in seconds. The tool is built on VEED’s industry-leading video technology and powered by Gemini models through Vertex AI. 
• Writer is a full-stack generative AI platform, running entirely on Google Cloud, that provides enterprises with a robust set of AI components, including its own Palmyra family of LLMs. In addition, Writer is now available to customers on Google Cloud Marketplace, making it very easy for enterprises to deploy the full-stack generative AI platform on Google Cloud.

Partnering with NVIDIA to accelerate AI development and deployment for startups

In addition to providing its AI platform and tools to the next generation of AI startups, Google Cloud also offers startups a comprehensive choice of AI-optimized infrastructure, including access to powerful GPUs through our partnership with NVIDIA. For example, Google Cloud was the first cloud provider to offer the G2 VM powered by NVIDIA L4 Tensor Core GPUs, purpose-built for generative AI inference workloads and graphics. We also offer the A3 VM powered by NVIDIA H100 Tensor Core GPUs, which are purpose-built to train and serve demanding generative AI workloads. AI startups need fast and reliable access to this technology to perform both AI model training and inference.

To provide more startups with fast-tracked access to this technology, we’re launching a new partnership between the Google for Startups Cloud Program and NVIDIA Inception. NVIDIA Inception is a free program for startups that offers its more than 18,000 members technology expertise, NVIDIA Deep Learning Institute Course credits, access to NVIDIA hardware and software, and other benefits. This new partnership will provide an accelerated path to use Google Cloud infrastructure and access to Google Cloud credits for early-stage members using NVIDIA GPUs, as well as software for testing or production on Google Cloud. We are also offering new go-to-market and product support for Inception members and fast-tracked onboarding to Google Cloud Marketplace. 

Additionally, we’ll include NVIDIA’s developer education content in Google’s AI Startup school training series, and Google for Startups Cloud Program members will have the opportunity to participate in NVIDIA Inception Capital Connect to support their fundraising efforts.

Announcing our inaugural cohort: AI-First North American Startups

Finally, to help AI-focused startups scale quickly while building responsibly, we’re thrilled to introduce the Google for Startups Accelerator: AI-First. This new program is for startups building AI solutions based in the U.S. and Canada. This is the first of several of these kinds of AI-focused programs we'll offer throughout the year in Europe, India, and Brazil.

This equity-free program provides 10 weeks of hands-on mentorship and technical project support to startups using AI in their core services or products. Selected startups will collaborate with a cohort of top peer founders and engage with leaders across Google. The curriculum will give founders access to the latest AI tools, and will also include workshops on tech and infrastructure, UX and product, growth, sales, leadership, and OKRs.

We’re thrilled to introduce the 15 AI startups selected for this accelerator: Aptori, Augmend, Backpack Healthcare, BrainLogic AI, Cicerai, CLIKA, Easel AI, Findly, Glass Health, Kodif, Liminal, mbue, Modulo Bio, Rocket Doctor, and Sibli.

The program kicks off this week at Next and culminates with a high profile Demo Day for potential partners, customers, and investors. After graduation, startups will join the dynamic Google for Startups Accelerator community, where they will receive ongoing support and have the opportunity to build lasting connections with like-minded founders, mentors, and investors.

We’re thrilled that Google Cloud has become the best place for startups to build with AI — with our broad choice of models and developer tooling, our open and integrated AI stack, and our expanding resources, support, and routes-to-market for early stage companies.

You can learn more about the Google for Startups Cloud program here.