Jaya Gupta and Ashu Garg’s recent reflection on context graphs is a timely reminder that the next major shift in AI will not come from bigger models alone, but from deeper understanding. The systems that win will be the ones that can operate in real-world complexity that span across more than one system or more than one organizational boundary, where meaning is not isolated in single data points but embedded in relationships, workflows, history, and intent.
A true context graph is not simply a retrieval layer or a technical construct, or even just a fancier RAG pipeline. It is a living representation of how an enterprise actually works. It captures not only entities, but the connections between them, the processes that shape them, and the trust mechanisms that govern them. Context graphs become powerful when they contain provenance, auditability, and semantic coherence, so that AI can reason over reality rather than fragmented information.
This is especially true in sustainability, where context is everything.
Sustainability data is inherently multi-dimensional. An emissions number is never just a number. It matters where it came from, what assumptions were used, how it maps to a framework, what organizational boundary it belongs to, and whether it can withstand regulatory and audit scrutiny. Sustainability is one of the most complex data environments enterprises have ever faced, because it spans internal operations, supply chains, shifting regulations, and stakeholder expectations all at once.
That is why Pulsora’s AI strategy places the trusted context graph at the center of its moat. Pulsora explicitly defines its strategy around “building a trusted context graph as a true moat” and grounding it in workflows and audit integrity, rather than treating context as an afterthought.
What makes Pulsora’s approach compelling is the richness of what sits inside that sustainability context graph, which is not just about emissions transactions. It connects organizational hierarchy, regulatory and disclosure metrics, audit history, workflow approvals and rejections, emission factors, transformations, estimations, and more into a unified sustainability intelligence layer.
Earlier in my career, I founded a startup that helped lead the creation of the Master Data Management category at Velosel, which is still TIBCO MDM. One of the deepest lessons from that experience was that intelligence always requires grounding. AI cannot reason effectively when reality is scattered across disconnected definitions of the same entities and metrics. Context has to be assembled deliberately.
But sustainability takes this idea further. The challenge is no longer only about consistency. It is about Significance.
This is where Pulsora adds relevance to the “Why,” not just the “How.”
Most systems in sustainability focus on the mechanics of reporting. They help companies calculate emissions, complete disclosures, and comply with requirements. That is the “how.” But the real transformation comes from understanding the “why.”
Why are emissions rising in one region but not another?
Why is a supplier emerging as an unexpected Scope 3 risk?
Why does one decarbonization pathway matter more strategically than another?
Why do certain risks compound across water, biodiversity, and physical exposure?
When a context graph contains the full fabric of sustainability meaning, AI can move beyond automation and into insight. Pulsora is designed to support that shift through end-to-end flows that span collecting, measuring, reporting, and analyzing sustainability performance, ultimately uncovering actionable insights that inform strategy rather than just compliance. The comprehensive nature of the application that spans a broad set of sustainability use cases including GHG, Data Collection, Regulatory and Voluntary Reporting lends itself to these use cases - naturally. It would be incredibly hard to do this with a narrower or niche application scope.
This is also why agents become truly valuable only when they are grounded in such a graph. Pulsora’s roadmap includes agents for data collection, anomaly detection, benchmarking, pre-audit readiness, and scenario modeling, all operating as intelligence layers on top of the sustainability context graph foundation.
The point is not that agents themselves are the moat. The moat is the depth of trusted context underneath them.
Ashu and Jaya are exactly right that context graphs represent the next evolution in AI. Sustainability is all about cross-organization, cross-systems, unstructured data, adhoc actions and is one of the clearest domains where this becomes undeniable, because sustainability is not just a data problem. It is a decision problem. It is a strategic problem. And increasingly, it is a planet-scale problem.
The future will belong to platforms that can help organizations answer not only “What did we report?” but also “Why does it matter, and what should we do next?”
That is the promise of a true sustainability context graph, and it is exactly what Pulsora is building.
