Tapestry, an energy software company backed by Google, has implemented its HyperQ AI platform to manage the initial site control review for PJM Interconnection’s first reformed interconnection queue cycle. In a blog post, the company stated that it had handled 213 generation applications totaling 22,026 GW of potential capacity in less than an hour. PJM is the largest Regional Transmission Organization (RTO) in the US, providing electricity to 13 states and the District of Columbia. The RTO recently transitioned from a first-come, first-served interconnection queue to a first-ready, first-served cluster study process in compliance with FERC Order 2023. The site control verification process checks whether applicants hold legal rights to the land they plan to develop and has become a mandatory requirement for applications to proceed. According to Tapestry, its HyperQ platform converted 4,581 raw site control documents into 2,328 agreement bundles and performed 133,312 parallel compliance assessments covering contract terms, chain-of-title continuity, exclusivity, and minimum acreage requirements. The median processing time per assessment was 6 minutes and 15 seconds, with 95% of tasks completed in under 28 minutes. The company stated that its process was 20 times faster than traditional manual review, although this comparison relies on what it calls a conservative lower-bound historical baseline of four hours per application. The HyperQ platform leverages multimodal AI to evaluate PDFs, maps, and legal documents, providing page-specific citations that PJM engineers use to make the final compliance decisions. Tapestry claims that the citation system has cut the amount of time engineers need to find supporting documentation. It added that it worked with PJM for ten months to embed the tariff requirements into the platform’s logic, leveraging a dataset of 234 historical queue applications to fine-tune its performance. The company reported that HyperQ reviewed 99 percent of applications in under an hour. Tapestry subsequently announced plans to expand the AI system to electrical data reviews—such as single-line diagrams and equipment specifications—and ultimately to a developer-facing tool that lets applicants pre-validate submissions before formal filing. The RTO signed a partnership agreement to implement Tapestry’s agentic AI platform last April. Tapestry is a Google X “moonshot” initiative that is using AI to develop what it describes as the world’s first unified model of the electrical grid.
