United StatesAI startup Mirendil has entered a multi-year partnership with Google Cloud, securing over $100 million in compute resources to advance its self-improving AI research. The collaboration grants access to Google's TPUs and Nvidia GPUs, enabling Mirendil to develop AI capable of iterative self-improvement. This aligns with broader industry trends of cloud providers supporting startups and AI firms competing for computational resources. The deal represents nearly half of Mirendil's recent $1 billion valuation seed round. The startup aims to automate scientific research across fields like medicine and materials science, drawing inspiration from Anthropic and other similar initiatives. Mirendil's co-founders argue that self-improving AI could emulate human learning processes, continuously refining its capabilities over time. They emphasize the importance of matching workloads to appropriate hardware, highlighting Google's flexible infrastructure as crucial for reducing costs and enhancing efficiency.
Bias read (Center): The article presents a factual report on a business development in the AI sector without overt ideological slant. It discusses technological advancements, corporate partnerships, and industry trends without taking a clear stance on political or social issues. The tone remains neutral, focusing on Mì
