Gruve raises $50 million to solve what its CEO calls AI's biggest problem: power (www.businessinsider.com)

Gruve, an infrastructure startup focused on AI inference, has secured $50 million in funding to address the critical power shortages affecting AI applications as the industry transitions from training to inference. With this latest investment, Gruve's total funding has reached $87.5 million. The company collaborates with data center providers to harness unused power and space, now accessing approximately 500 megawatts across major U.S. cities. CEO Tarun Raisoni emphasizes that the main obstacle in AI today is insufficient power availability, and Gruve is innovating to connect “stranded” power resources. The startup offers a more hands-on engineering approach compared to traditional cloud giants, catering specifically to companies that may lack in-house machine learning expertise. Gruve currently provides 30 megawatts of power available for order across multiple states, optimizing the geographic distribution of services to ensure quicker and more cost-effective data transmission. With a diverse customer base, including companies like PayPal and Stanford Health Care, Gruve plans to expand globally while focusing on hiring engineers and machine learning researchers to enhance its inferencing software, crucial for meeting the growing demands of AI workloads.