Kingsley Uyi Idehen
Newton, Massachusetts, United States
23K followers
500+ connections
23K followers
500+ connections
About
Founder & CEO, OpenLink Software.
OpenLink Software is an industry acclaimed…
Articles by Kingsley Uyi
-
Everyday Reading as Compounding Knowledge Infrastructure
Everyday Reading as Compounding Knowledge Infrastructure
Everyday reading has always been a knowledge process. The difference now is that an Agentic Web, buoyed by a Semantic…
1 Comment -
An Agentic Web powered by an increasingly Semantic Web has Arrived!
An Agentic Web powered by an increasingly Semantic Web has Arrived!
“…what to hold in your head and what to offload…” – Jaya Gupta from Experience is now a tax. Since the advent of…
9 Comments -
Using Attribute-Based Access Control (ABAC) to Power Economic Equilibrium in the Age of AI
Using Attribute-Based Access Control (ABAC) to Power Economic Equilibrium in the Age of AI
The Quiet Asymmetry Long story short: at OpenLink Software we've seen this movie before—it’s an age-old battle to…
13 Comments -
Building on LLM Wikis: Turn Any Document URL into a Living, Queryable Semantic Web Knowledge Graph with One AI Agent Skill
Building on LLM Wikis: Turn Any Document URL into a Living, Queryable Semantic Web Knowledge Graph with One AI Agent Skill
Last weekend, Andrej Karpathy shared a compelling idea: LLM Wikis. Feed in documents.
3 Comments -
Language, Context, and Digital Notebooks in the age of AI
Language, Context, and Digital Notebooks in the age of AI
Here’s a short note—in video form—that sheds light on the growing importance of machine-computable data, information…
4 Comments -
Why Does Virtuoso Exist?
Why Does Virtuoso Exist?
We created Virtuoso to address the challenges associated with semantic harmonization of data spaces — spanning…
5 Comments -
Unlocking the AI-Driven Economy: The Power of Modular Agent Skills
Unlocking the AI-Driven Economy: The Power of Modular Agent Skills
As the founder and CEO of OpenLink Software, I’ve spent decades bridging data silos with technologies from the Semantic…
2 Comments -
Unlocking Economic Value in AI: The Role of Agent Skills and Loosely Coupled Data Spaces
Unlocking Economic Value in AI: The Role of Agent Skills and Loosely Coupled Data Spaces
The AI revolution isn’t just about bigger models or faster inference — it’s about modularity, interoperability, and…
1 Comment -
Dealing with SQL’s Ubiquity in the Age of AI: Unearthing Semantically Rich Context Through Interoperability, Not Replacement
Dealing with SQL’s Ubiquity in the Age of AI: Unearthing Semantically Rich Context Through Interoperability, Not Replacement
In a data-driven world increasingly hungry for interconnected insights, one language refuses to yield its throne: SQL…
-
Use Case Exercise -- Knowledge Graph Generation from existing Relational CSV Project
Use Case Exercise -- Knowledge Graph Generation from existing Relational CSV Project
When teams say they need a “Knowledge Graph,” what they usually mean is a production system capable of ingesting messy…
20 Comments
Activity
23K followers
-
Kingsley Uyi Idehen shared thisHere's another case of notes generation and persistence to a knowledge graph based on an article of interest. In this case, the State of Martech 2026 research report by Scott Brinker and Frans Riemersma. Links: https://lnkd.in/eKxmRJZb Just click to explore. The whole thing (HTML and underlying Knowledge Graph) have been generated by an AI Agent leveraging loosely coupled skills for handling the transformations from PDF to RDF. Explainer: https://lnkd.in/e3sYNjSj #Martech #LinkedData #SemanticWeb #KnowledgeGraphs #HowTo #CMO #CDO #CTO #CxO #AgenticWeb #SmartAgent #AIKingsley Uyi Idehen shared thisThe full State of Martech 2026 report is here. 🙌 And it’s free. Ungated. Attached directly to this post as a PDF. No landing page. No form to fill out. No “leave a comment and we’ll DM you” hoop. Frans Riemersma and I have spent the past 6 months analyzing where martech is really headed — across the 15,505 products in the 2026 Marketing Technology Landscape, our survey of 200+ marketing leaders, and a deep dive into 70 AI use cases reshaping marketing. A few of the big stories inside: 1️⃣ The martech landscape is effectively flat for the first time in 15 years — but underneath that headline, the market is anything but static. 1,488 products were added. 1,367 were removed. We reveal which categories are growing, shrinking, or churning out the old with the new. 2️⃣ It took the martech landscape 15 years to reach 15,000 products. The MCP server ecosystem hit nearly twice that in 18 months. Anthropic, OpenAI, Google, and Microsoft all converged on the same open protocol — companies that agree on roughly nothing else. We unpack what this means for the integration physics that have shaped martech for two decades, and where agentic gravity is pulling the stack next. 3️⃣ The 20-year debate between “consolidate on a suite” and “best-of-breed” has a 2026 answer: neither. The stack is stratifying. AI-native tools are winning the creation layer. Incumbent SaaS still owns much of the orchestration layer, where workflows and data already live. And homegrown AI is showing up where proprietary data, business logic, brand experience, or customer context really matter. Across our AI use case data, leading teams aren’t choosing between AI in existing SaaS, new AI-native tools, or homegrown solutions. They’re often using all three — sometimes for the same use case. 4️⃣ AI doesn’t make martech complexity disappear. It exposes it. When a human is in the loop, they can quietly compensate for bad handoffs, messy data, disconnected systems, outdated product information, and fuzzy governance. When an AI agent is orchestrating a customer interaction, answering a buyer’s question, recommending a product, or triggering a workflow, your internal architecture leaks directly into the customer experience. We dig into what context engineering means for marketing ops in this new reality — and why the inner game now shapes the outer game more than ever. Huge thanks to GrowthLoop, Hightouch, Knak, MoEngage, Pegasystems, Progress Software, and SAS for sponsoring this research and making it possible for us to share it freely with the community. 🙏 Download the PDF below. Read it. Share it. Challenge it. Give us your hot takes. The chrysalis stage of martech is messy — but something beautiful is taking shape. 🦋 #marketing #martech #AI #MartechDay
-
Kingsley Uyi Idehen shared thisYep! An important question that we must ask ourselves. Fundamentally, we cannot allow the “devil in the software default” to erode our uniquely human skills. Web 2.0 attacked our attention as part of its myopic attempt to monetize Internet and Web connectivity. We cannot allow that dysfunction to expand into the Agentic Web now rapidly taking shape. The counterbalance to AI lies in refusing to abdicate uniquely human capabilities that cannot be replicated in software—i.e., not succumbing to the allure of the “devil in the default” in the name of convenience.Kingsley Uyi Idehen shared thisNobody talks about what AI is actually doing to your attention span… and after 13 years and 200+ enterprise AI deployments, it is the most important conversation we are not having. 🧠 We have spent so much energy debating whether AI will replace jobs that we missed what it is already quietly taking from us. The ability to focus. The tolerance for hard problems. The confidence to trust our own judgment before consulting a tool. I call it Intellectual Atrophy… and it shows up in boardrooms and everyday decisions far more than anyone admits. The 7 signs it is already affecting yours: 1️⃣ You reach for AI before attempting to think through the problem yourself 2️⃣ You struggle to hold focus through anything longer than a few paragraphs 3️⃣ You feel uncomfortable when you cannot immediately prompt an answer 4️⃣ You have stopped forming your own opinion before searching for one 5️⃣ You cannot remember the last time you solved something complex entirely alone 6️⃣ You accept AI output without scrutiny because verifying feels like too much effort 7️⃣ Sitting with ambiguity for more than a few minutes has become genuinely difficult The brain is a muscle. And the most sophisticated technology ever built is currently doing your reps for you. Outsource your tasks. Never your thinking. Which one hit closest to home? #ai #leadership #intellectualatrophy #futureofwork #artificialintelligence #criticalthinking #teamhuman #cognitivehealth
-
Kingsley Uyi Idehen shared thisEveryday reading has always been a knowledge process. The difference now is that an Agentic Web, buoyed by a Semantic Web, makes that process computable. Check out my new post that explains how. #CDO #CAIO #CIO #CTO #CMO #CxO #AgenticWeb #Agentic #SmartAgent #SemanticWeb #KnowledgeGraphs #LinkedData #Explainer #HowTo #FileSystem #DataSpaces #VirtuosoRDBMSEveryday Reading as Compounding Knowledge InfrastructureEveryday Reading as Compounding Knowledge Infrastructure
-
Kingsley Uyi Idehen shared thisSpot on! Note: AI is fundamentally just software equipped with a long-sought multimodal and multilingual natural language interface. Take the term “artificial intelligence” literally at all times.Kingsley Uyi Idehen shared thisThe single most useful sentence Sol Rashidi, MBA gave me in our hour together: "Make sure AI happens with you and not to you." — Sol Rashidi That sentence is a diagnostic question disguised as a principle. If AI is happening to you, someone else is deciding how your role gets reshaped. Probably an Excel spreadsheet. Probably built — in Sol's words — by "consultants who don't have to provide for your family the way you do." If AI is happening with you, you're at the table. You're the one mapping which parts of your work get augmented, which get outsourced to a model, and which stay in your hands because they require judgment a model can't replicate. The shift is happening either way. The only choice is whether you're the architect or the input. Three things to put on this week's calendar: 1. List the ten tasks you do most often. Mark which a model could do tomorrow. 2. For the ones you marked, ask: what would I redesign so this becomes my leverage instead of my exposure? 3. Have that conversation with your manager before someone has it about you. You don't need a strategy deck to do any of these. You need an hour with a notebook. Full episode: https://lnkd.in/g_ySN6sP Checkout: Cyera an EVOTEK Production Sol's site: https://www.solrashidi.com Sol on LinkedIn: https://lnkd.in/dEMX5itM Sol's book — Your AI Survival Guide (Wiley, 2024): https://lnkd.in/d55hTQpP #AILeadership #FutureOfWork #AI #Career #DigitalIntelligence
-
Kingsley Uyi Idehen shared thisA really important post for all Africans—both on the continent and in the diaspora. #Africa #LODLAM #GLAM #Culture #OpenData #LinkedDataLibraries as AI Infrastructure: The African EditionLibraries as AI Infrastructure: The African Edition
-
Kingsley Uyi Idehen shared thisWe already blend SPARQL, GraphQL, and SQL—so why not add OpenCypher? This reinforces the point that RDF, as an abstract language for structured data definition, is not tied to any specific declarative query language. The real issue—often obscured by endless distractions—is breaking down data, information, and knowledge silos. This is achieved powerfully by leveraging hyperlinks as super keys equipped with two fundamental capabilities: 1. Unambiguous naming 2. Boundless access to data, information, and knowledge by reference That’s what truly puts the “Web” in the Semantic Web vision—and why a Web connecting documents (i.e., Web 1.0) changed the world. #CDO #CAIO #CIO #CTO #CMO #CxO #KnowledgeGraphs #SemanticWeb #LinkedData #OpenCypher #SPARQL #SQL #GraphQL #VirtuosoRDBMSKingsley Uyi Idehen shared thisGraph data has reached an interesting moment. The industry has matured beyond a simple debate between RDF graphs and Labeled property graphs. The real question now is: How do we combine semantic interoperability with developer-friendly graph querying? That is where the idea of a Semantic Property Graph, or SPG, becomes useful. We are previewing openCypher support in Virtuoso that demonstrates a simple but important point: a labeled property graph can be modeled in RDF without changing the rules of RDF. The practical benefit is bigger than query syntax. It means LPG-style nodes and relationships can be grounded in hyperlinks as identifiers, and not only local IDs trapped inside one database. Not only opaque handles that need a separate mapping layer. Actual dereferenceable, shareable identifiers that can participate in a broader Web of Data. That brings semantic behavior to property graph workflows: - Globally meaningful identifiers - Navigability across graph boundaries - Vocabulary reuse through prefixes and IRIs - Named graph scope - SPARQL interoperability - Inference and semantic integration - Linked Data publication and discovery In this model: - Nodes are RDF resources. - Relationships are RDF predicates. - Property keys can be RDF predicates. - Relationship properties are represented under the hood using RDF reification, with RDF 1.2 support as the future direction. The same graph remains 100% RDF, while becoming accessible through an openCypher-style LPG interface, SPARQL, and even SQL interfaces. This is not about replacing SPARQL or RDF. It is about enhancing Virtuoso with openCypher as another query-language option on top of SPARQL for querying RDF-backed graphs. This also matters for neuro-symbolic AI. LPG-style traversal is useful for graph exploration, feature discovery, and application development, while RDF adds machine-computable semantics, identity, vocabulary alignment, inference, and explainable reasoning to the same graph. From an LPG lens, reasoning becomes a way to enrich a graph with inferred relationships, classifications, and constraints without forcing developers out of a property graph interaction model. This is the point: RDF and labeled property graphs are not mutually exclusive camps. They are different interaction models over graph-shaped data. Virtuoso openCypher is about reducing the impedance mismatch between them, so graph applications, semantic data, analytics, AI systems, and integration workloads can meet in the same system. The either/or argument is overdue for retirement. Here are some live examples: - Federated query against Wikidata: https://lnkd.in/ekD3rTbb - Federated query against DBpedia: https://lnkd.in/eJ54-s45 - Relationship Property syntax that adheres to RDF standards : https://lnkd.in/euiuAxAp #VirtuosoRDBMS #OpenCypher #SPARQL #RDF #KnowledgeGraphs #GraphDatabases #LinkedData #SemanticWeb #NeurosymbolicAI
-
Kingsley Uyi Idehen shared thisIn a nutshell.Kingsley Uyi Idehen shared thisI am one of those whose research points to the conclusion that the mind is computable. Sentience is an exquisite consequence of the laws of physics. I see no evidence that requires the supernatural; I find panpsychism to be a god-of-the-gaps fantasy; I see no requirement for any quantum pixie dust. What I do see is that evolution has led the cosmos through billions of years of experiments and mistakes and failures, eventually bringing forth artifacts with intelligence, with consciousness, with sentience, with life itself, by growing and combining very simple things then morphing them into extraordinarily complex ones. An underlying challenge in all this discourse, of course, is that these words represent ineffable concepts that human language strains at and fails to define crisply, and so we end up talking past one another, with emotion rather than rational and informed dialog. That notwithstanding, it is the ultimate hubris to conclude that only we humans can be conscious or sentient: a multitude of creatures living among us possess those properties to varying degrees (and we already treat multitudes of them with questionable ethical consideration). It is hubris to demand that the intelligence and consciousness and sentience we experience is its only form. It is hubris for we humans to assert that at this moment are at the cusp of building sentient artifacts. My understanding and experience suggests that while some individual elements are within our understanding, there are many others we do not know that we even know we need to know. Moreover, there exists a problem of systems architecture at extreme scale, and that we are a long way from understanding how to engineer that. But I predict that some day we shall. This is not that day, nor is it a day in the lives of any of this present generation. I expect that embodiment will be necessary (all contemporary approaches are deeply impoverished with regard to building things in and of this world, with sensing and acting now only a tiny fraction of what organics experience in all its noise and ambiguity). I posit that in this journey, we will co-evolve, compelled to reconsider what it means to be human. I know that we will find value from the things we create, but I also know that there exist fundamental and unavoidable risks that will harm us. I fear most the wealthy and privileged few who assert they alone know the way and should be trusted to bring humanity to this new digital promised land; I fear the organizations and nations who seek to use these things for greater power and control. I suspect that some strive to build artificial super intelligence (because they want to be gods) while others reject the possibility (because they cherish our uniqueness). And yet, I am hopeful, and confident in the resilience of the human sprit to endure, and in the end, be better for the journey. What an extraordinary time to be alive.
-
Kingsley Uyi Idehen shared thisChanged the world, and its implications are still being unpacked!Kingsley Uyi Idehen shared thisOn this day, 33 years ago, CERN put the World Wide Web software in the public domain. Proposed by Sir Tim Berners-Lee, the web was originally created to allow scientists and institutes working on CERN data to share information accurately and quickly from across the globe. Later, CERN made a release available with an open licence, to maximise its dissemination. Here we see: 1️⃣+2️⃣ The document that officially put the World Wide Web into the public domain 3️⃣ Tim Berners-Lee’s NeXT computer that served as the original web server and development machine for the WWW project 4️⃣ A screenshot of the recreated page of the first website Find out more about the history of the #WWW: https://lnkd.in/dm8qc2zy
-
Kingsley Uyi Idehen shared thisHere’s a Semantic Web vision presentation generated using NotebookLM from my notes over the years on this important endeavor—building on the Internet connectivity unleashed by the original Web, which achieved mass adoption and acclaim. Naturally, it’s also inspired by a growing number of posts celebrating the Semantic Web vision and the power of ontologies (once a “banned” word!) as we enter the age of AI, powered by AI agents and skills loosely coupled with data spaces (databases, knowledge bases, filesystems, and APIs). Regarding the deck, it also shows current NotebookLM limitation regarding slide corrections. I gave up by the 16th try regarding the hallucinated content of slide #4 😃 #CDO #CAIO #CIO #CTO #CMO #CxO #SemanticWeb #KnowledgeGraphs #Explainer #Software
Kingsley Uyi Idehen replied to a comment
8h
OpenAI Codex AI Agent, GPT-5.5 LLM, and the same AI agent skills:
1. HTML-based Infographic (notes summary) -- https://linkeddata.uriburner.com/DAV/demos/daas/state-of-martech-2026-gpt5-chat-1.html
2. Knowledge Graph Overview via Faceted Search & Browse Page -- https://linkeddata.uriburner.com/describe/?url=https%3A%2F%2Fcontent.martechday.com%2Fstate-of-martech-2026.pdf%23analysis&distinct=1
Kingsley Uyi Idehen commented on a post
8h
Some additional links showcasing different AI agents, AI agent skills, and large language models (LLMs).
Claude Code AI Agent, DeepSeek-v4 LLM, and the same AI agent skills:1. HTML-based Infographic (notes summary) -- https://linkeddata.uriburner.com/DAV/demos//daas/state-of-martech-2026-deepseek_v4pro-1.html
2. Knowledge Graph Overview via Faceted Search & Browse Page -- https://linkeddata.uriburner.com/describe/?url=https%3A%2F%2Fcontent.martechday.com%2Fstate-of-martech-2026.pdf%23article&distinct=1
Kingsley Uyi Idehen replied to a comment
12h
Scott,
Here's what I promised:
https://linkeddata.uriburner.com/DAV/demos/daas/state-of-martech-2026-minimax_m2.5free-1.html
Just click to explore. The whole thing (HTML and underlying Knowledge Graph) have been generated by an AI Agent and loosely coupled skills for handling the transformations from PDF to RDF.
Explainer:
https://linkeddata.uriburner.com/DAV/demos//daas/state-of-martech-2026-minimax_m2.5free-1.html#about
#HowTo #SemanticWeb #KnowledgeGraphs #LinkedData #AgenticWeb
-
Kingsley Uyi Idehen liked thisKingsley Uyi Idehen liked thisThe question I'm thinking about the most in AI isn't how big the opportunity is, rather, it's how long it will take to diffuse into the enterprise. My take on a recent Startup Grind panel with Yuri Sagalov, Christine Tsai, and Tomasz Tunguz: longer than most people think. The answer matters because: 1. Jobs - which roles get reshaped, augmented, or eliminated 2. Productivity - the timing of AIs impact per worker 3. Capex - trillions in data center spend needs enterprise revenue to pencil out 4. Startups - how aggressively to invest and what revenue ramps are realistic So why slower AI diffusion than we think? * Executive credit wars. Execs jockey for ownership, slowing AI decisions. * Software choices = workforce choices. Headcount equals power. * Change management is work. Rethinking approvals, handoffs, etc is hard. * Legacy tech stacks. Pre-cloud systems can't expose the APIs, data, etc. * Fragmented data. Critical data is distributed across many silos. * Tribal knowledge. Most valuable processes lives in people's heads. * Governance gaps. Security, audit, and access controls need to emerge. I'm bullish on enterprise AI, but diffusion will be slower - and messier - than most expect. We may overestimate the near-term adoption curve, yet may still underestimate the long-term impact.
-
Kingsley Uyi Idehen reacted on thisKingsley Uyi Idehen reacted on thisBusy week this week - the Knowledge Graph Conference kicks off at Cornell University in New York and SHACL starts to flex its muscles. --------------------------------------------- AI+Semantics NewsBytes — 6 May 2026 --------------------------------------------- 👉 KNOWLEDGE GRAPH CONFERENCE 2026: THE FIELD ARRIVES AT CRITICAL INFRASTRUCTURE 👉 LOVELACE LAUNCHES ELEMENTAL: CAN A STARTUP OUT-GRAPH PALANTIR? 👉 ONTOLOGY-CONSTRAINED NEURAL REASONING: EMPIRICAL PROOF FROM PRODUCTION 👉 SHACL-DS: EXTENDING VALIDATION INTO THE NAMED-GRAPH LAYER 👉 MCP BECOMES THE OPENAPI FOR AGENTIC AI — AND SEMANTIC ARCHITECTS SHOULD CARE 👉 xpSHACL: WHEN VALIDATION ERRORS NEED EXPLANATIONS, NOT JUST REPORTS
Experience
-
-
-
-
-
-
-
Projects
-
RDF 1.1
- Present
Revise and update the RDF specifications, without adversely affecting deployment. (Note: this is impossible, so we had to compromise.)
Other creators -
-
DBpedia
- Present
Linked Open Data rendition of Wikipedia. Basically, Modern, Web-scale and structured query friendly database driven by OpenLink Virtuoso.
Other creators -
-
Linked Open Data Cloud (LOD)
- Present
Massive Big Data collective comprised of datasets from a variety of domains such as: general knowledge (Wikidpedia), Life Sciences (Bio2RDF), Media (BBC), Government (Data.Gov and Data.Gov.UK), and many others. This massive collection of data is accessible via HTTP based Data Source Names and query languages such as SPARQL and SQL.
Other creators -
Languages
-
Edo (Bini)
Native or bilingual proficiency
-
English
Native or bilingual proficiency
Recommendations received
4 people have recommended Kingsley Uyi
Join now to viewOther similar profiles
Explore more posts
-
Mike Cornell
👀 Worth a quick watch! If you’re an enterprise with a legacy Databricks + Snowflake implementation ( ya know the one where you do ETL in Databricks and copy the resulting “gold” datasets to Snowflake for BI; or the one where you store data in Snowflake and copy data to Delta/Iceberg for Databricks to do ML? ) 🛑STOP! It’s time to modernize! Store your data once in open formats that you own! Secure and govern it one place! Read-from/write-to it from BOTH Databricks and Snowflake securely and efficiently! 💪
-
Sunil Jagani
MySQL Downtime Is Costing You More System failures don’t send warnings—they strike when you least expect them. The real question is: Is your MySQL high availability ecosystem truly prepared? In this practical guide, AllianceTek breaks down how MySQL HA works in real-world scenarios, highlighting common failure points and how to design systems that stay online when it matters most. Don’t wait for downtime to expose vulnerabilities. Build resilience into your infrastructure today. Read more: https://lnkd.in/g4X35jra #MySQL #HighAvailability #DatabaseManagement #ITInfrastructure #TechLeadership
-
Kalos Consulting
Matthew Monroe and ♛Paul Mitchell will be presenting practical guidance on how to get up and running with FileMaker Pro and local AI models, addressing questions like when to use LLM vs. when to use RAG, what AI models to choose, and whether to use SQL or native Find functionality. They’ll be exploring the latest features and applying them to solve real-world use cases. Matt Monroe has been the owner of Harmonic Data since 2004. He and his wife Debbie have two children. Paul Mitchell is a Low Code Engineer at Harmonic, with over 27 years’ experience working with FileMaker Pro. He and his wife Karie have five children. They are determined to make these new technologies accessible to a wider audience, and this will be a presentation aimed at both business owners and developers, covering not only the how, but the why. Claris, an Apple company #Kalos #BGL #SmallBusinessConference. #Filemaker
4 Comments
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content