\n<\/p>\n
Today Microsoft is updating its startup program to include a free Azure AI infrastructure option for high-end GPU virtual machine clusters, for use in training and running large language models and other deep learning models.<\/p>\n
Y Combinator<\/a> (YC) and its community of startup innovators will be the first to access this offering in private preview to a limited cohort. YC has an unmatched reputation as a pioneering startup accelerator helping launch transformative companies including Airbnb, Coinbase and Stripe. Now YC startups will have the technical resources they need to quickly prototype and bring to market cutting-edge AI innovations. Our close collaboration with YC provides valuable insights into the infrastructure needs of early-stage AI companies, ensuring our offering delivers optimal value to additional startups going forward.<\/p>\n \u201cWith the overwhelming infrastructure requirements needed to do AI at scale, we believe that providing startups with high-performance capabilities tailored for demanding AI workloads will empower our startups to ship faster,\u201d said Michael Seibel, Managing Director of Y Combinator.<\/p>\n<\/blockquote>\n We are also working with M12<\/a>, Microsoft\u2019s venture fund, and the startups in M12\u2019s portfolio which will gain access to these dedicated supercomputing resources to further empower their AI innovations.\u00a0 Over time, our vision is to partner with additional startup investors and accelerators, with a goal of working with the ecosystem to lower the barrier to training and running AI models for any promising startup.<\/p>\n Microsoft Azure offers cloud-based scalable AI infrastructure, built for and with the world\u2019s most sophisticated AI workloads, from delivering the largest and most complex AI models including GPT-4 and ChatGPT through Azure OpenAI Service to developers to infuse AI capabilities into many apps. Azure AI infrastructure is fueling groundbreaking innovations. Infrastructure requirements to do AI at scale are often overwhelming, but with Azure\u2019s global infrastructure of AI-accelerated server offerings with networked graphics processing units (GPUs), startups building advanced AI systems will be able to leverage these high-performance capabilities to accelerate innovation.<\/p>\n On top of world-class infrastructure, we will also provide tools to simplify deployment and management through Azure Machine Learning. This enables easy low-code or code-based training of custom models and fine-tuning of frontier and open-source models, simplified deployment and optimizations like Low Rank Adaptation, DeepSpeed and ONNX Runtime (ORT).\u00a0 Further, startups can deploy AI solutions with peace of mind knowing all deployments are secure and backed by Microsoft\u2019s principles for Responsible AI.<\/p>\n Empowering startups to build transformative solutions powered by AI<\/strong><\/p>\n AI is transforming industries and startups are leading that innovation, creating new business and societal value quicker than many thought possible. According to a recent KPMG survey<\/a>, the near-term demand is real, with 75% of U.S. CEOs stating that generative AI is a top investment priority, 83% anticipating an increase in generative AI investment by more than 50% in the next year, and 45% saying investment will at least double. For startups, this represents a once-in-a-generation opportunity to bring groundbreaking impact to a market hungry for change.<\/p>\n To help startups meet this opportunity, last year we introduced Microsoft for Startups Founders Hub<\/a> \u2013 designed to help founders speed development with free access to GitHub and the Microsoft Cloud as well as unique benefits including free access to $2,500 of OpenAI credits to experiment and up to $150,000 in Azure credits that startups can apply to Azure OpenAI Service. Startups also receive 1:1 advice from Microsoft AI experts to help guide implementation. The Microsoft Pegasus Program<\/a>, an extension of Founders Hub, links enterprise customers with startup solutions for immediate deployment. Seventy-five percent of Pegasus startups have landed deals with Fortune 1000 companies via increased reach across Azure Marketplace.<\/p>\n Startups using Azure AI to develop cutting-edge solutions for today\u2019s problems<\/strong><\/p>\n Whether you have a product in market or just an idea, Microsoft provides startups with the tools they need to rapidly build and scale AI solutions. Already, we are seeing the results of empowering startups to innovate with AI to improve customer support, detect and address health conditions and advance immersive gaming experiences. Here are just a few examples of the cutting-edge innovation happening now:<\/p>\n Commerce.AI dramatically increases call center productivity with Azure OpenAI Service<\/u><\/strong> Inworld: The next-generation AI character engine for immersive gaming BeeKeeperAI<\/u><\/strong> is helping catch rare childhood conditions early YouTube Video<\/p>\n\n
Commerce.AI<\/a> uses Azure OpenAI Service and Azure AI Services to make call centers more efficient. Azure Cognitive Services uses a Commerce.AI model to transcribe interactions in real time, including into multiple languages. After the call, Azure OpenAI Service creates a summary with customer contact information, topics of conversation and embedded sentiment analysis. The system selects next steps and follow-up action items from pre-generated options, and the customer service agent exports the information to Microsoft Dynamics 365 in one quick step.<\/p>\n
<\/u><\/strong>Inworld<\/a>, a Silicon Valley startup, is a fully integrated character engine that goes beyond language models to give users complete control over AI non-player characters (NPCs). With Inworld, users can customize their characters\u2019 knowledge, memory, personality, emotions and narrative role. Inworld uses Azure AI technologies like Azure OpenAI Service to power its advanced natural language understanding and generation.<\/p>\n
<\/u><\/strong>AI tooling company BeeKeeperAI<\/a> enables AI algorithms to run in a private and compliant way in healthcare environments.\u00a0The company is pioneering an effort to leverage confidential computing to train an algorithm for predicting a rare childhood condition using real patient health records. By encrypting both the data and algorithm and using Microsoft Azure\u2019s confidential computing, the company has enabled the algorithm to analyze identifiable health information in a secure, sightless manner.<\/p>\n