Business Technology Roundtable

Every enterprise racing to deploy Agentic AI is making an infrastructure decision it may not recognize as one. When an AI agent connects to internal systems and proprietary content, the protocol governing that connection determines whether the enterprise controls the terms of access or cedes them by default. Model Context Protocol (MCP) has become the standard linking agents to backend data and tools, and it is spreading through corporate technology stacks faster than the governance frameworks meant to secure it.  For CIOs and CISOs, this stopped being a developer tooling question months ago. It is now a sovereignty question: who defines the boundary between an AI agent and a company's proprietary intelligence. The MCP Adoption Security Gap Gartner projects that by 2028, 25 percent of all enterprise Generative AI (GenAI) applications will experience at least five minor security incidents per year, up from 9 percent in 2025. That trajectory tracks closely with how fast MCP has been ...

Applied-AI adoption inside enterprises has nearly tripled in four years, moving from roughly 7 percent of OECD firms in 2021 to 20 percent in 2025. Over the same stretch of time, the share of the workforce demonstrating recognized AI skills has barely cleared 1 percent. That gap, not the pace of AI model releases, is the strategic problem every C-suite now has to own and purposefully resolve. The Widening Gap Between Deployment and Capability Half of small and medium enterprises now cite a lack of skills as the reason they have not adopted generative AI tools, according to the OECD's newly declassified paper, Skills in the AI Age. Across 10 member countries, roughly one third of job vacancies already sit in occupations with high AI exposure, a share expected to keep climbing as generative tools embed themselves in everyday workplace software. Executives who treat this as a training line item rather than a workforce architecture problem will spend the next few years chasing a target...

A strategic business case and global market assessment The enterprise artificial intelligence landscape is undergoing a profound structural transition, particularly with the Professional Services related industries. As AI foundation models commoditize basic text generation and public datasets reach cognitive density limits, forward-thinking organizations are realizing that generic intelligence offers no sustainable competitive advantage. In response, enterprise architects and strategic advisors are shifting focus toward architectures that capture, structure, and monetize internal institutional knowledge. Central to this structural pivot is the concept of the Proprietary Intelligence Foundry. The foundry represents a strategic pipeline framework designed to transform an enterprise's undocumented human judgment into a permanent, defensible, and sovereign digital asset. Conceptual Foundations The Proprietary Intelligence Foundry is defined as a strategic pipeline architecture that cap...

In June, I wrote about why AI training had moved inside the enterprise : the open web has run dry of the cognitive density frontier models now need, and the next phase of the Applied-AI race would be won on access to human tacit knowledge, not public data. I said the leading tech vendors had already restructured to capture it. What I didn't have yet was a single, undeniable proof point of a professional services firm making the same bet with its own money. Now I do. Kirkland & Ellis, the world's highest-grossing law firm, is spending roughly $500 million of its own revenue over the next three to four years building proprietary AI systems it will own outright. A meaningful piece of that is a co-built platform with Palantir called the Fund Formation Engine, designed to run private-equity fund documentation, side-letter drafting, and obligation tracking across Kirkland's fund-formation practice. Kirkland didn't pick Palantir for its model. It picked Palantir for its on...

Google just published the most detailed empirical picture yet of how Generative AI is actually being used across the global economy, and it should reset how executive teams talk about AI transformation. The AI & Economy ATLAS study, built on 15 million de-identified interactions across the Gemini App, Google AI Mode, and the Gemini API, maps usage against more than 800 occupations, 4,000 work tasks, and 150 countries. The headline finding is not that Applied-AI is transforming work. It is that AI has diffused everywhere while penetrating almost nowhere near as deeply as the enterprise boardroom narrative suggests. The AI Coverage Illusion Gemini usage now touches 68 percent of detailed occupations, covering occupations that represent 88.4 percent of total U.S. employment. That is the statistic executives will quote. It is also the one most likely to mislead them. Among occupations with any measurable AI usage, the median worker is applying AI to just 21 percent of their constituen...

The Applied-AI Requirement, Seen from My Own Desk A few months ago, I began evaluating a paid license from some of the leading AI providers on the assumption that my advisory work would eventually outgrow the free AI app tiers. It has not happened yet. Each time I approached the point of subscribing, a new free or entry-level release arrived that adequately covered what I actually needed: drafting, research synthesis, and editorial refinement.  My requirements were never exotic. They were representative of a large share of knowledge work, which is precisely the point. This is not a story about frugality. It is a story about an AI capability moving target. The capability that once justified a premium license a year ago is now embedded in the free tier of the same provider, or matched by a competitor's low-cost model. Independent benchmark trackers have shown the performance gap between open and proprietary models narrowing from double digits to less than one percentage point within ...

The public internet, long treated as an inexhaustible resource for training large language models, has run dry. Not in terms of raw volume, but in terms of the cognitive density that frontier AI now requires. Research I published through GeoActive Group's Applied-AI Initiative confirms what a growing number of senior researchers have quietly acknowledged: the next phase of the AI race is being won or lost on access to human tacit knowledge, and the leading tech vendors have already restructured their organizations to capture it. This is not an incremental refinement to existing AI training methodology. It is a wholesale reorientation of how the most resource-intensive companies in the world are deploying their most valuable internal asset: the unwritten reasoning of their best people. The Structural Bottleneck Driving This Shift Three converging constraints have forced this strategic pivot. First, models trained on generic web content have hit a reasoning ceiling. They perform ade...

Economic growth is fueled by strategic investment. Venture capital (VC) has quietly become one of the most consequential forces in the global networked economy, yet most C-suite leaders engage with it only at the margins. That needs to change. A landmark new report from the World Economic Forum and Stanford Graduate School of Business, released this month, offers a comprehensive and at times sobering assessment of where the VC industry stands and where it is headed. For senior executives navigating technology strategy, capital allocation, and competitive positioning, the findings carry direct implications. The Scale of the Opportunity and the Strain Beneath It VC assets under management have grown more than sixfold since 2008, reaching $3.4 trillion globally in 2025. Seven of the ten largest companies in the world by market capitalization, including Apple, NVIDIA, and Amazon, received venture backing in their early stages. Among U.S. public companies founded in the past 50 years, VC-ba...