VMware Private AI
Private AI is an AI architectural approach that aims to balance the business gains from AI with the practical privacy and compliance needs of the organization. VMware AI Labs has worked across VMware R&D and with our industry partners to create a platform for AI services that ensures privacy, control and choice while maintaining quick time-to-value.
Explore Innovations in Private AI
Discover how VMware accelerates customers’ generative AI initiatives with more secure, ubiquitous and affordable Private AI solutions.
Reference Architecture on VMware Cloud Foundation (VCF) 5
The VMware GenAI Baseline Reference Architecture provides organizations with the fastest path from ML project to production.
Visit our GitHub repository to see code samples of our Reference Architecture.
Spin Up Clusters in Seconds Using Open Source Ray
Ray on VMware Cloud Foundation enables Data Scientists and MLOps engineers to scale AI and Python workloads effortlessly by utilizing their current compute footprints for ML workloads instead of defaulting to the public cloud. Through a new VMware Cloud Foundation plug-in, VMware will deliver Ray’s ease of automated AI workload deployment and scaling to Cloud Foundation environments.
Fine Tuning StarCoder
LLMs as coding assistants for developers is becoming mainstream in the industry. However, for companies with strict data privacy and security requirements, we need an on-premise solution. To solve this challenge, VMware is collaborating with Hugging Face to implement their open source code LLM: StarCoder on-premises.
Unlocking Tomorrow's Intelligence
AI Labs is dedicated to shaping a more intelligent future through its incubation projects and cutting-edge AI research, driving the evolution of AI within VMware and beyond.
Project Trinidad
Project Trinidad detects zero-day attacks in Modern Applications by applying Machine Learning to layer 7 East-West API traffic. Traffic is collected in kernel using eBPF. Our Machine Learning models infer normal application behavior from this traffic, and can detect anomalies that are symptomatic of threats and attacks.
VMware NSX+ Intelligent Assist
AI-driven security assistant for swift, precise threat assessment and remediation featuring auto-processing of NSX+ NDR alerts, identifying familiar campaigns, and recommending actions. This chatbot uses generative AI and LLMs to aid analysts in natural language queries and explanations during triage, functioning as an expert collaborator in security decisions.
Confidential Computing
Confidential Computing offers guest applications end-to-end secure enclave support on the VMware vSphere platform. We designed generic enclave execution control and attestation/certifications APIs that make minimal assumptions on the underlying hardware capabilities. Looking toward the future, we are building a partner ecosystem to help accelerate the adoption of confidential computing while positioning VMware as a leader in multi-cloud security.
Research in AI
The VMware Research Group (VRG) works on trustworthy, private and resource efficient AI/ML and the underlying systems and algorithms that make AI possible. VRG creates unique differentiation for VMware’s products and advances the state of the field through impact on the research community.
Learn more about what’s new.
Learn more about what’s new.
DoCoFL: Downlink Compression for Cross-Device Federated Learning
DoCoFL offers a comprehensive and novel downlink framework for cross-device Federated Learning that significantly reduces the downlink bandwidth requirements without degrading ML performance.
Advancing Research in the Memory Sub-System
Memory management is a complex and intricate area in operating systems. Explore our advances in memory sub-systems that impact the performance of existing and emerging workloads.
Research Publications
AI Labs Research brings university and industry researchers together to push the boundaries of technology. Explore our diverse set of recent publications to learn more about our work in AI and Systems.
Responsible AI
As AI/ML technologies see increasing adoption in important societal applications, it is critical to develop and deploy these techniques responsibly with a clear set of guiding ethical principles.
Unleashing the Power of AI with Ethical Integrity
At VMware, our adoption of ethical AI principles must align with both our EPIC2 values and our ESG goals to build trustworthy and safe AI systems, which respect privacy, security and content ownership policies, and behave in a fair and transparent manner.
