Perspectives
Developed positions on why AI programmes stall, what has to be redesigned, who controls intelligence once it is distributed, and who captures the value. Every perspective is published in full, with an ungated PDF.
Start here
Perspective 2 · Why it fails · 17 min read
Why AI Transformation Fails
AI programmes do not fail because the models are weak. They fail because the organisation was never redesigned to absorb what AI does to decision authority.
Read the perspective →
Where AI programmes break, and why the break is structural.
Perspective 1
Organisations approach AI through tools, but AI value depends on structure, incentives and control, so buying the technology without redesigning the operating model reliably produces adoption without results.
Perspective 5
When AI absorbs most of a function's volume, the work that remains needs a different structure, and organisations that bank the margin while keeping the old structure are compounding the wrong thing.
The operating model, architecture and data decisions AI forces.
Perspective 3
An AI operating model is designed or it is inherited: an organisation that does not decide, decision by decision, where intelligence lives and who controls it will have that answer set by procurement, cost metrics and its first incident.
Perspective 8
AI makes the data model, not the application, the durable enterprise asset: organisations that keep relocating process-era applications will compound drag, while those that own their data model and place inference by decision will compound advantage.
Perspective 10
AI scales only when roughly seven parts of the effort go into governance, controls and authority structures for every one part spent on new execution capability. Enterprises that invert the ratio buy capability they cannot absorb.
Authority, accountability and governance once intelligence is distributed.
Perspective 4
Joint ventures are run by their operating system, not their cap table: whoever controls decision rights, systems, data and people controls the venture, and AI moves that control into the platform one partner usually owns.
Perspective 6
Hallucinations are not defects that the next model will patch out. They are pressure responses, and organisations contain them by borrowing governance from psychiatry and psychotherapy, not from software quality assurance.
Perspective 11
Structural transformation led by the people whose authority rests on the legacy will stall; it needs an outsider with explicit authority over the operating model and over personnel, for a defined term, who then leaves.
Edge, cloud, platforms and partners: who ends up controlling enterprise intelligence.
Perspective 7
When inference runs on site, the decision runs on site, and most enterprises have moved their decisions to the edge without moving the decision rights that should govern them.
Perspective 9
Telcos and hyperscalers are fighting over the same control point, data access at the edge, and enterprises that buy edge infrastructure on unit price will hand that control to whichever supplier arrives first.
Perspective 12
Frontier models now reach every integrator through every cloud, and the labs and clouds are building delivery arms of their own, so an enterprise should choose whom to trust with its AI estate on accountability and independence, never on model access.
These are related positions rather than an ordered series; each stands on its own and links to the others where the argument continues.
A perspective may be developed from signals, trends and operator notes, or commissioned directly. It may begin with a pattern in the evidence, an operating problem or a question deliberately commissioned; signals, trends and notes can inform it without becoming mandatory stages.
What matters is the quality of the argument: the sources supporting it, the counterevidence it can withstand and the operational consequences it explains. Editorial review determines whether it is ready to publish.
Suggest a Perspective
Perspectives are one layer of BdG Advisory's Intelligence work — alongside Signals, Trends and Operator Notes. Signals are observations. Trends link them into derived intelligence. Operator Notes and Perspectives are content artifacts with provenance — a note may be developing thinking, a perspective is a stated position, and either can draw on the others without having to.