Source URL: https://cloud.google.com/blog/topics/startups/google-for-startups-program-fuels-more-ai-innovation/
Source: Cloud Blog
Title: The Google for Startups Cloud Program fuels more AI innovation
Feedly Summary: Startups are building the future of AI models, infrastructure, and applications on Google Cloud.
These fast-growing AI startups want to be able to accelerate their product development, ensure capacity, and scale cost-effectively. And they can with many of our unique offerings: a broad choice of Google-built, open, and partner models; our Vertex AI platform; access to both TPUs and GPUs; and highly capable data infrastructure.
We’re committed to making Google Cloud the best place for startups to utilize and build AI. This includes making it frictionless for startups to get up and running with our technology, such as providing access to the best developer tools via our Vertex AI platform. We’ve also taken a “model-neutral” stance that provides greater choice for customers, so they always have the best models suited to their particular tasks. And there’s Google’s own experience building and scaling AI, which we readily share with customers through our collaborative approach.
We’ll share updates in these areas and more at our Startup Summit on October 8th.
Many startups begin working with us in their early stages, through our partnerships with leading accelerators like Y Combinator, AI2 Incubator, and the Entrepreneurs Roundtable Accelerator. We also have our own Google for Startups Cloud Program, which provides up to $350,000 of credits for AI startups over two years, in addition to technical training, business support, and Google-wide offers — which enable startups to build quickly and scale effectively.
As we near our annual Startup Summit, we’re taking the opportunity to recognize AI startups that have recently joined the Google for Startups Cloud Program, with each applying AI in innovative ways. They’re training specialized models in areas like video creation using TPUs or deploying AI on the frontlines of medical research. They’re helping developers be more productive, improving software security, and much more — all on Google Cloud’s platform.
Startups that have recently joined the program are:
Higgsfield AI, focused on AI video generation for social media, utilizes Vertex AI and GPUs to train and deliver its platform for generative AI videos.
Sakana.ai, a Tokyo-based startup that leverages NVIDIA GPUs on Google Cloud, along with services like Filestore and Persistent Disk, to power its research activities and host its foundational models.
Form Bio, accelerates medical research and drug discovery with AI. They built their platform natively on Google Cloud using products like Vertex AI, BigQuery, and Compute Engine.
LiveX AI, which offers products including LiveX AI Agent and ChurnControl, is using products like Google Kubernetes Engine (GKE) to scale its platform cost-effectively — including reducing its operational costs by roughly 66%.
bhsn.ai, is building AI-powered software to help businesses improve contract processing, reviews, and research. They use a mix of services and products on Google Cloud including NVIDIA GPUs, GKE, and Cloud Storage.
Turbopuffer, is a serverless vector database built completely around Google Cloud Storage, which enables companies building around AI to store their vector embeddings without needing to manage servers.
Martian created an LLM router on Google Cloud to dynamically direct traffic between multiple models for cost savings, improved performance, and simplified AI.
General Agents, an AI startup training foundation models for general computer control.
LastMile AI, a platform that helps enterprise developers safely and confidently deploy AI applications in production by using Google Cloud and Vertex AI to rigorously measure and improve their performance.
Pioneers continuing to grow on Google Cloud
These customers join more than half of the world’s generative AI startups who are developing and scaling their products on Google Cloud, including:
Magic: With a mission to pave the way toward safe artificial general intelligence, Magic seeks to automate software engineering and research through frontier code models. The company uses Google Cloud’s AI Hypercomputer and NVIDIA H100 Tensor Core GPUs to build supercomputers that can achieve 160 exaflops of computing performance.
Mistral: Developer of open-weight models designed to create a safe artificial general intelligence that benefits all users, Mistral offers customizable and compute-efficient AI models leveraging Google Cloud infrastructure. Its models are available via Vertex AI on Model Garden.
Chainguard: Launched by a group of experts with backgrounds in open source software and software supply chain security, Chainguard delivers zero-CVE container images that provide secure-by-default infrastructure that prevents workflow friction and delights CISOs and developers alike. The company uses Google Cloud’s Serverless, BigQuery and Cloud Workstations to power its innovative platform, enabling rapid analysis of software supply chain data and facilitating secure, collaborative development workflows.
Pinecone: Pinecone creates infrastructure loved by developers for building accurate, secure, and scalable AI applications. With Pinecone serverless on Google Cloud, companies can easily ground GenAI apps in their proprietary data for AI search, RAG, and other use cases. Since first launching Pinecone serverless, over 30,000 organizations have collectively indexed more than 25 billion embeddings to accelerate GenAI prototyping and production.
Learn more about Google Cloud’s commitments to startups
Be sure to check out the Startup Summit to hear how Google Cloud is committing more resources to founders in the ways that matter most: accelerating growth, providing physical spaces that nurture community and innovation, enhancing skills and knowledge development, and leading the way in generative AI.
At the event, which takes place on October 8, you’ll hear from founders of AI startups like Vellum, Reve AI, and Click Therapeutics; VCs like Capital G and Lightspeed Venture Partners; plus accelerators, partners, and Google executives including our own CEO, Thomas Kurian. Speakers will discuss the latest generative AI advances and the challenges shaping the current startup climate, capped off with plenty of advice so you can tap into the opportunities we’re seeing today.
AI Summary and Description: Yes
**Summary:** The text highlights Google’s initiatives to support AI startups on its cloud platform, detailing their tools, resources, and collaborative models. It emphasizes Google Cloud’s versatility and focus on promoting innovation in AI applications, while showcasing specific startups that leverage their technology for diverse use cases.
**Detailed Description:**
– *Google Cloud as a Catalyst for AI Startups*: The text emphasizes Google’s commitment to supporting AI startups by providing innovative infrastructure like the Vertex AI platform, TPUs, and GPUs for accelerating product development.
– *Model-Neutral Approach*: Google Cloud adopts a model-neutral stance to allow startups the flexibility to choose the most appropriate AI models for their tasks, fostering innovation and customization.
– *Collaboration and Support*: Various partnerships with leading accelerators (e.g., Y Combinator) enhance Google’s connection with startups. The Google for Startups Cloud Program offers financial credits, technical training, and business support to emerging companies.
– *Key Startups Highlighted*:
– **Higgsfield AI**: Focuses on AI-generated videos for social media, utilizing Google’s infrastructure for enhanced video production.
– **Sakana.ai**: A research-focused startup in Tokyo using NVIDIA GPUs for foundational model hosting.
– **Form Bio**: Accelerates drug discovery processes through AI models built on Google Cloud.
– **LiveX AI**: Offers cost-effective scaling solutions using Google Kubernetes Engine (GKE).
– **Martian**: Created a dynamic LLM router for optimizing model traffic, showcasing advanced use of cloud capabilities.
– *Infrastructure for Generative AI*: The document underscores that over half of the world’s generative AI startups utilize Google Cloud, highlighting companies like Magic and Chaosguard that leverage Google Cloud for secure and efficient AI solutions.
– *Upcoming Startup Summit*: Scheduled for October 8, the summit aims to provide insights into Google Cloud’s commitments to startups, featuring speakers from AI startups and venture capitalists, addressing challenges, opportunities, and advances in generative AI.
This text is significant for professionals in AI, cloud computing, and security sectors as it showcases a growing ecosystem of support for AI innovation and addresses potential compliance and security concerns associated with deploying AI applications at scale.