LangGrant Launches the Industry’s First Open Standards Initiative for a Safe Enterprise
BELLEVUE, Wash. – LangGrant (earlier known as Windocks), a leader in database modernization and synthetic data, today announced the Enterprise Reasoning Initiative, a fully open source project supported by 10 innovators across the AI ecosystem. The Initiative’s goal is to develop an open interoperability standard that all vendors, enterprises, and developers can use that makes human expertise a persistent part of how enterprises reason with AI.
This creates a direct safety mechanism for autonomous Enterprise AI: human judgment, enterprise policies, evidence, and approvals can become part of the reasoning that precedes consequential decisions and actions, while the reasoning itself remains available afterward for accountability, audit, and improvement.
This unique approach will solve four key AI issues organizations face today:
● Safety — As AI agents become more autonomous, human judgment, enterprise policies, and approvals can be applied to the reasoning behind consequential decisions and actions—not just reviewed after the fact.
● Transparency – The initiative would eliminate the “black box” most frontier models use by default for problem-solving and prevent AI systems from reinventing new reasoning every time they complete a task.
● Availability – This open source standard would be available for implementation by anyone.
● Reusability – The project aims to create a standard to make the reasoning they apply in their AI implementations a durable object that can be reused and managed over time.
“The future of Enterprise AI should make human expertise more valuable. When an expert improves AI reasoning, that expertise should become part of the organization’s intelligence, and not disappear when the conversation ends. That’s how people and AI learn together,” said Ramesh Parameswaran, CEO of LangGrant and co-founder of the Enterprise Reasoning Initiative. “As AI models become more capable and autonomous, enterprises face a choice. They can increasingly hand analysis and decision-making to AI, with people primarily reviewing the results. Or they can build systems in which human expertise and AI reasoning continuously build on each other. The Enterprise Reasoning Initiative is designed for the second path.”
The initiative proposes a common representation for reasoning that allows people, software tools, and AI models to work collaboratively on the same reasoning over time. Human judgment – changing an assumption, applying domain expertise, rejecting an inference, introducing a business rule, or refining a definition – becomes part of the reasoning that AI and other people can build upon the next time.
From human-in-the-loop to a shared learning loop
Most human-in-the-loop approaches involve people reviewing or approving what AI has already produced. Enterprise Reasoning proposes going further: human judgment becomes part of the reasoning itself.
Reasoning is represented as a durable enterprise artifact containing the sources, steps, evidence, human and AI contributions, semantics, policies, versions, and approvals behind an analysis or decision. Humans can then review, modify, re-execute, and reuse that reasoning collaboratively using different software tools. That reasoning can then be reviewed, modified, executed again, and reused collaboratively by humans using different people, software tooling, and different AI models.
Instead of beginning every interaction with an empty prompt, enterprises can begin with well-documented, standardized reasoning that their team members and AI have already developed together. The result is that human expertise becomes reusable organizational intelligence rather than disappearing after each AI interaction.
“We’ve seen this interoperability problem before. With each major computing transition, the industry eventually needs common standards that let different products work together,” said Bob Kruger, chief product officer of Almaden AI, and co-founder of the Enterprise Reasoning Initiative. “AI creates a new challenge: people and AI systems need a common way to build on reasoning over time. Enterprise Reasoning is an effort to establish that common foundation.”
An open standard for reasoning, not another AI model
Enterprise Reasoning does not standardize how AI models reason or prescribe how applications should implement AI. It standardizes the representation of reasoning so people, software tools, and AI systems can exchange, validate, explain, and build upon it.
The proposed standard is initially focused on six capabilities: structured reasoning; human judgment throughout the reasoning process; reasoning lifecycle management; reasoning across multiple enterprise information sources; progressively evolving semantic intelligence; and attribution of reasoning and decisions to business outcomes. This makes reasoning a first-class enterprise artifact – much like source code, documents, or datasets – rather than disposable AI output.
Enterprise Reasoning is intended to complement emerging AI interoperability standards. Model Context Protocol (MCP) addresses how models access tools and context. Agent2Agent (A2A) addresses communication between agents. Enterprise Reasoning addresses another layer: how the reasoning shared among people, enterprise software, and AI models can be represented, preserved, and continuously improved.
A common data format alone is not enough. Interoperability requires a shared information model so software systems understand the meaning of reasoning artifacts, not simply their syntax.
10 companies support Enterprise Reasoning
LangGrant is launching the Enterprise Reasoning Initiative with support from multiple AI innovators. Initial supporters include:
● Almaden AI
● Causal Dynamic Labs
● Conflux
● DeepGraph
● GirardAI
● LEIT Data
● Ngentix
● Proof Analytics
● Skyhook
● The Knowledge Graph Guys
The initiative is seeking additional participation from enterprise software vendors, AI companies, enterprise architects, executives with deep industry and domain expertise, researchers, and standards organizations. Participants will contribute to the development and review of the Enterprise Reasoning information model, reference implementations, interoperability testing, domain semantics, and the evolution of the standard.
A detailed white paper outlining the Initiative’s goals, along with an initial standards work plan, is available. Interested parties can find out more about how to participate by visiting the website or contacting the Initiative.
The post LangGrant Launches the Industry’s First Open Standards Initiative for a Safe Enterprise appeared first on SD Times.
Tech Developers
No comments