Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallEnterprise semantics is the work of giving business terms and data relationships shared, machine-readable meaning. In an October 1, 2026, InfoWorld opinion article, Suresh Srinivas argues that AI could make this long-running ambition more practical—not by putting more company data into a model, but by giving data agents context about what the data means, which sources to trust, and what people have already learned. The proposal is promising, but the article’s reported performance figures are not independently validated benchmarks.
What enterprise semantics means in practice
Imagine a leader asks, “How many customers do we serve in Europe?” or a CFO requests revenue by customer segment. These sound like straightforward questions, but an organization may not have one shared definition of “customer,” “Europe,” “revenue,” or “segment.” Teams can use different rules, tables, or time periods and each produce a plausible answer.
Enterprise semantics is the effort to make those concepts and their connections explicit across the business and its data. The goal is not merely to let an AI system search databases. It is to help an agent interpret a question, identify relevant data, apply agreed definitions and rules, and explain which information supports its answer.
Srinivas, identified by InfoWorld as co-founder and CEO of Collate and the OpenMetadata open-source project and a former chief architect of Uber’s data platform, makes this case as an industry practitioner. His article is an argument for an approach, not an independent evaluation of semantic platforms or AI data agents.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →#1 Best Overall
Why earlier approaches were difficult to sustain
Formal meaning took specialist effort
Earlier Semantic Web work pursued a compelling goal: represent information in a form machines could interpret. Standards such as RDF, OWL, and SKOS persisted, but building enterprise ontologies around them could be expensive. The work often required people who understood both business concepts and technical modeling, lengthy workshops, and ongoing manual upkeep as systems and definitions changed.
As Srinivas puts it, “The Semantic Web had the right vision and the wrong tools.” That is his assessment of why the vision struggled to scale, rather than evidence that semantic standards themselves have no value.
Rank #2
Glossaries align people, but may not give machines enough structure
A data-catalog glossary can help employees use a metric name consistently. A written definition alone, however, may not tell an agent which entities the term involves, how those entities relate, which rules apply, or which dataset is authoritative. The distinction is between documenting a definition for people and connecting concepts, properties, relationships, and rules in a form that supports machine reasoning.
The three kinds of context in the proposal
Srinivas describes an AI-ready context layer as three connected kinds of information. They answer different questions, so one does not replace the others.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →| Context type | What it helps answer | Examples of information |
|---|---|---|
| Data context | What data exists and how usable or connected is it? | Schemas, data-quality signals, lineage, and usage information |
| Semantic context | What do business concepts mean, and how do they relate? | Ontologies, business relationships, and rules—for example, how “customer” relates to “net revenue” |
| Memory context | What has the organization already corrected or learned? | A shared, persistent record of feedback, corrections, and organizational knowledge |
In Srinivas’s framing, a language model still needs help understanding structured data and business authority. “LLMs still need to be told what the structured data means, how business concepts are defined, and which data is authoritative.” The three contexts are intended to supply that help: discovery, meaning, and accumulated guidance.
What AI could change—and what still needs people
The proposed change is that AI may assist with the labor-intensive parts of building and maintaining context. It could help populate technical metadata, draft ontology elements for experts to review, and flag possible drift when business definitions or data change. That could reduce the effort needed to create a useful context layer compared with relying entirely on manual discovery and upkeep.
This is not a claim that an agent can autonomously establish correct business definitions. A draft ontology can encode a mistaken assumption; a metadata signal can be incomplete; and a stored correction may not apply in a new situation. People still need to decide what terms mean, determine which data is authoritative, review proposed changes, and govern access and use. The article does not demonstrate that these workflows are universally reliable or eliminate governance work.
How strong is the evidence for better answers or lower cost?
The article reports two figures as results from Srinivas’s company’s internal tests: answers were “seven times more accurate” and query workloads were 86% lower. It does not give the test design, sample size, baseline, or independent replication. Treat them as company-reported results, not general expectations for enterprise deployments.
It also reports a Gartner prediction that organizations prioritizing semantics in AI-ready data could see AI costs 60% lower by 2027. The InfoWorld article does not link the underlying Gartner report, so the forecast cannot be assessed in detail from that article alone. It is a reported prediction, not a measured outcome for a typical organization.
These numbers point to questions worth testing, but they do not establish that semantics alone caused the reported changes, or that another company would see the same results. Any evaluation should measure answer correctness and workload against a defined baseline, and include the effort and cost of creating, reviewing, and maintaining context.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.How to judge an enterprise-semantics approach
The argument suggests practical questions for evaluating an approach, without establishing that any particular product meets them:
- Meaning: Can it represent business entities, relationships, and rules, or does it mainly store text definitions?
- Data coverage: Does it connect schemas with quality, lineage, and usage signals?
- Memory: Can it preserve expert corrections and make them available to future tasks?
- Maintenance: How are changes in data structures and business definitions detected, reviewed, and applied?
- Governance: Who approves definitions, determines authoritative sources, and reviews machine-generated context?
- Evidence: Are answer quality, workload, and total cost measured against a clear baseline under conditions relevant to your organization?
Srinivas’s larger claim is that organizational knowledge could accumulate instead of being rebuilt or lost between projects. Whether that happens depends on the quality of the context, the review process, and how reliably changes reach the systems and people that need them—not simply on adding an AI agent.
Free tools Windows power users keep installed
One-click scans. No signup required.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
