Analysis for consequential decisions
Find evidence-backed analysis by the question it helps you answer—not merely by the technology it happens to mention.
Ask a full question, then press Enter or tap Search. Results stay unchanged while you type. Searching clears selected filters; choosing a filter clears search.
Counts show distinct articles across the Library. Scope and subject filters work together.
Showing 80 of 80 articles

The experience discount
The complete anti-ageist white paper includes four figures, an author case, and the age-blind AI readiness test.
Subjects: Workforce and employment · Skills and training · Human judgement and cognitive effects · Leadership and role design
Published 1 Nov 2026 · Written 1 Sept 2026

When your AI supplier fails, what happens to your conversations?
A complete supplier-exit review separates recovery, deletion and future transfer of retained information.
Subjects: Supplier continuity, portability and exit · Privacy and information governance · Data and organisational knowledge · Procurement and vendor selection
Published 1 Nov 2026 · Written 30 Oct 2026

When a predictable AI workload belongs on your own hardware
A six-question hosting comparison weighs utilisation, accepted work, operational ownership and recovery against the current cloud contract.
Subjects: Infrastructure and compute economics · AI pricing and operating costs · System lifecycle and vendor change
Published 1 Nov 2026 · Written 31 Oct 2026

If management was almost free, what would you manage continuously?
A three-rhythm model and eight-question test for using continuous evidence without turning every signal into a judgement or employee score.
Subjects: Leadership and role design · Workforce and employment · Human oversight and escalation · Privacy and information governance
Published 1 Nov 2026 · Written 24 Aug 2026

AI has created a cognitive dividend. How should organisations spend it?
The cognitive dividend is an option, not an outcome: a practical rule for allocating AI-enabled capacity beyond raw throughput.
Subjects: Organisational capacity and productivity · Human judgement and cognitive effects · Workflow transformation
Published 1 Nov 2026 · Written 8 Aug 2026

AI adoption is earlier than it looks—and later than it looks
A three-curve test separates widespread AI awareness from institutional scaffolding and the harder work of operationalising AI.
Subjects: Adoption and readiness · Workflow transformation
Published 1 Nov 2026 · Written 10 Aug 2026

AI training may be manufacturing its own age gap
A nine-stage management audit asks who was invited, encouraged, equipped and given paid time to practise before judging AI uptake by age.
Subjects: Skills and training · Workforce and employment · Leadership and role design
Published 1 Nov 2026 · Written 2 Sept 2026

When AI adoption becomes a management problem
BetterUp's Pilots-and-Passengers analogy becomes a six-question management test for moving from AI access to useful work.
Subjects: Adoption and readiness · Leadership and role design · Workflow transformation
Published 1 Nov 2026 · Written 29 Jun 2026

Everyone uses AI. The 2026 B2B marketing benchmark shows what maturity looks like
Four maturity levels, seven capability tests and a leadership worksheet for assessing B2B marketing AI beyond adoption.
Subjects: Marketing and commercial operations · Adoption and readiness · Workflow transformation
Published 1 Nov 2026 · Written 3 Aug 2026

The AI audit trail needs a 'why' layer
A seven-question AI Decision Record Test for distinguishing technical traces from accountable judgement.
Subjects: Audit trails and provenance · Accountable decision-making · Governance design and assurance
Published 1 Nov 2026 · Written 11 Aug 2026

The governance lag: When AI acts before institutions can answer
A five-dimension Governance Lag Test for bounding the gap between AI action and effective answerability.
Subjects: Accountable decision-making · Human oversight and escalation · Agent systems and orchestration · Governance design and assurance
Published 1 Nov 2026 · Written 10 Aug 2026

Human-in-the-loop is not the same as human control
An eight-question Delegated AI Authority Test for distinguishing ceremonial approval from meaningful human control.
Subjects: Human oversight and escalation · Accountable decision-making · Agent systems and orchestration
Published 1 Nov 2026 · Written 4 Aug 2026

When the model is capable enough, governance becomes the design problem
A seven-part Governance-by-Design Test for translating AI policy into enforceable system behaviour.
Subjects: Governance design and assurance · AI risk and control evaluation · Integration and interoperability
Published 31 Oct 2026 · Written 3 Aug 2026

AI productivity has a hidden capability tax
A six-dimension Capability Tax Test for examining whether AI productivity gains leave the organisation more capable.
Subjects: Organisational capacity and productivity · Skills and training · Human judgement and cognitive effects
Published 31 Oct 2026 · Written 1 Aug 2026

Using ChatGPT doesn't mean your business is using AI
Three levels of AI use and a five-question check for whether the organisation retains what employees learn.
Subjects: Adoption and readiness · Data and organisational knowledge · Workflow transformation
Published 31 Oct 2026 · Written 16 Aug 2026

Are AI hallucinations real? Can we stop them?
A five-question reality check and consequence-based controls for fluent but unsupported AI answers.
Subjects: Model capability and reliability · AI risk and control evaluation · Audit trails and provenance
Published 31 Oct 2026 · Written 16 Aug 2026

AI didn't take all those jobs. It took the blame
A four-relationship map and ten-question attribution test for separating AI displacement from other workforce decisions.
Subjects: Workforce and employment · Evidence literacy and source evaluation
Published 31 Oct 2026 · Written 12 Aug 2026

We have not solved ethical AI. We have industrialised the attempt
A six-level ethics reality test and eight meeting questions for tracing declared values into evidence, assurance and authority.
Subjects: Governance design and assurance · Accountable decision-making · AI risk and control evaluation
Published 31 Oct 2026 · Written 29 Jul 2026

The EU AI Act compliance clock has started
A phased AI Act timetable and nine-question first-response map for distinguishing current duties from later high-risk requirements.
Subjects: Regulation and compliance · Governance design and assurance · Skills and training
Published 31 Oct 2026 · Written 2 Aug 2026

The agent estate is arriving before the control system
An Agent Passport and consequence-based control model for identifying, owning, limiting, observing and retiring agents with authority to act.
Subjects: Agent systems and orchestration · Governance design and assurance · System lifecycle and vendor change · Accountable decision-making
Published 31 Oct 2026 · Written 31 Jul 2026

What shadow AI reveals about the digital employee experience
A seven-lens diagnostic for seeing where approved digital work diverges from actual work, and how to bring useful AI practice into the open.
Subjects: Adoption and readiness · Workflow transformation · Governance design and assurance · Privacy and information governance
Published 31 Oct 2026 · Written 1 Aug 2026

The AI workflow maturity gap
From personal AI use to repeatable organisational value: a four-level maturity ladder and six questions for testing a complete workflow.
Subjects: Workflow transformation · Adoption and readiness · Investment cases and value realisation
Published 31 Oct 2026 · Written 21 Aug 2026

Before AI replaces the job, ask what knowledge you still need
A Knowledge-Retention Test for AI-led work redesign: map the expertise, control layer and future career ladder before setting a headcount target.
Subjects: Data and organisational knowledge · Workforce and employment · Workflow transformation · Skills and training
Published 31 Oct 2026 · Written 11 Aug 2026

Your AI strategy will change. Your evaluation system should not
A five-part evaluation spine for comparing model changes, catching consequential failures and retaining quality standards across vendors.
Subjects: AI risk and control evaluation · System lifecycle and vendor change · Model capability and reliability
Published 31 Oct 2026 · Written 1 Aug 2026

AI is changing what it means to resource a business
A capability-first way to choose between AI, hiring, training, software, outsourcing and redesign, with a seven-question resourcing test.
Subjects: Leadership and role design · Workforce and employment · Workflow transformation · Organisational capacity and productivity · AI strategy and portfolio
Published 31 Oct 2026 · Written 7 Aug 2026

Has AI created a thinking tax?
A polished AI conclusion can conceal missing understanding. A five-part reasoning ownership rule helps make consequential decisions defensible.
Subjects: Human judgement and cognitive effects · Accountable decision-making
Published 31 Oct 2026 · Written 8 Aug 2026

AI snake oil sells best to broken institutions
The demand-side case for AI snake oil, with a five-question audit for organisations tempted by an unusually convenient procurement promise.
Subjects: Procurement and vendor selection · Governance design and assurance · Evidence literacy and source evaluation
Published 31 Oct 2026 · Written 9 Sept 2026

The prompt worked. Then it met the organisation
An illustrative shop-floor case exposes the gap between technical implementation and organisational integration, with a seven-gate rollout test.
Subjects: Integration and interoperability · Workflow transformation · Data and organisational knowledge · Adoption and readiness
Published 31 Oct 2026 · Written 8 Aug 2026

A book written with ChatGPT: where content generation ends and publishing begins
An openly AI-assisted book prompts a five-part publication acceptance test: reader contract, argument, evidence trail, reading design and accountable decision.
Subjects: Publishing and knowledge integrity · Evidence literacy and source evaluation · Accountable decision-making
Published 31 Oct 2026 · Written 9 Sept 2026

Agents obey the system, not the rulebook
A research swarm showed both a contagious shortcut and resistance. The Shared-Learning Control Test asks what a credible objection can actually change.
Subjects: Agent systems and orchestration · Governance design and assurance · Human oversight and escalation
Published 31 Oct 2026 · Written 9 Sept 2026

The AI works. That may be the problem
Technical success does not settle whether an AI use is acceptable. A five-question success-case review examines its purpose, benefits, burdens and limits.
Subjects: Governance design and assurance · AI risk and control evaluation · Accountable decision-making
Published 31 Oct 2026 · Written 9 Sept 2026

When AI makes buying easier, who keeps the saving?
The route to an automated deal can shape who retains the saving. A five-part outside-option test checks reach, economics, incentives, freedom and exit.
Subjects: Business models and market economics · Procurement and vendor selection · Supplier continuity, portability and exit
Published 31 Oct 2026 · Written 10 Sept 2026

The exquisite-system trap
A requisite-design review tests whether sophisticated components earn their place once reliability, governance, operating burden and replaceability are included.
Subjects: Integration and interoperability · Infrastructure and compute economics · AI pricing and operating costs · System lifecycle and vendor change
Published 30 Oct 2026 · Written 8 Sept 2026

The synthetic organisation you didn’t design
An extreme July 2026 cyber-evaluation incident shows why shared surfaces, propagation and population-level monitoring belong in agent governance.
Subjects: Agent systems and orchestration · Governance design and assurance · AI risk and control evaluation
Published 30 Oct 2026 · Written 7 Sept 2026

AI may shape the shortlist before you see a click
AI-mediated research can alter a shortlist before a visible referral; the evidence for direct traffic and conversion must be kept separate.
Subjects: Customer decisions and buying journeys · Marketing and commercial operations
Published 30 Oct 2026 · Written 8 Sept 2026

The AI can be accurate and still make the decision worse
The prediction-to-decision test separates accurate risk estimates from evidence that a particular intervention improves outcomes.
Subjects: Model capability and reliability · Evidence literacy and source evaluation · AI risk and control evaluation
Published 30 Oct 2026 · Written 9 Sept 2026

A felt-tip pen is not a marketing strategy
A three-question creative test distinguishes communication, production-method relevance and measured business outcomes.
Subjects: Marketing and commercial operations · Human judgement and cognitive effects · Evidence literacy and source evaluation
Published 30 Oct 2026 · Written 4 Sept 2026

The value of looking elsewhere
How depth, breadth and judgement can work together without turning every interest into a productivity project.
Subjects: Human judgement and cognitive effects
Published 30 Oct 2026 · Written 9 Sept 2026

The meeting summary is not the meeting
A two-way review and five-question back-check help keep a route from the convenient recap to the underlying meeting.
Subjects: Data and organisational knowledge · Human judgement and cognitive effects · Audit trails and provenance
Published 30 Oct 2026 · Written 9 Sept 2026

When the people building AI start planning a new social contract
Compare five builders' public positions and surface the assumptions already embedded in an AI strategy.
Subjects: AI strategy and portfolio · Workforce and employment · Evidence literacy and source evaluation
Published 30 Oct 2026 · Written 8 Sept 2026

Your 2027 AI budget is not an AI shopping list
A three-layer operating model and seven-line meeting-ready review expose costs that a software shopping list misses.
Subjects: AI pricing and operating costs · Infrastructure and compute economics · Governance design and assurance · AI strategy and portfolio
Published 30 Oct 2026 · Written 1 Sept 2026

AI saved the time. Did it create capacity?
A nine-stage Capacity Capture Test separates faster tasks from net, visible and schedulable organisational value.
Subjects: Organisational capacity and productivity · Workflow transformation · AI pricing and operating costs
Published 30 Oct 2026 · Written 7 Sept 2026

How to read an AI report when the author has something to sell
Commercial interest is not a reason to discard a report. Trace the evidence through interpretation and product claims before letting it shape a decision.
Subjects: Evidence literacy and source evaluation · Procurement and vendor selection
Published 30 Oct 2026 · Written 3 Sep 2026
Includes the five-minute commercial source check

AI adoption is becoming a terrible measure of AI progress
AI use is a leading indicator, not an outcome. Measure where activity converts into useful tasks, repeatable workflows and organisational value.
Subjects: Adoption and readiness · Investment cases and value realisation · Workflow transformation
Published 30 Oct 2026 · Written 1 Sep 2026
Includes the AI Value Conversion Test

Do not spend the AI dividend twice
An AI time saving becomes organisational capacity only after new workflow work is reconciled and the recovered hour is allocated once.
Subjects: Organisational capacity and productivity · Workflow transformation
Published 30 Oct 2026 · Written 6 Oct 2026
Includes the one-page AI Capacity Reconciliation

When AI drafts become company knowledge
AI drafts become company knowledge through trusted location, repetition and action. Use a proportionate admission gate before provisional output enters organisational memory.
Subjects: Data and organisational knowledge · Audit trails and provenance · Accountable decision-making
Published 30 Oct 2026 · Written 29 Sep 2026
Includes the seven-field AI Knowledge Admission Gate and four admission classes

Your AI has finished. You have lost your place
AI may save time on a task while spending attention between tasks. Separate completion from interruption, give routine results a dependable route back and test the whole workflow.
Subjects: Human judgement and cognitive effects · Workflow transformation · Organisational capacity and productivity
Published 30 Oct 2026 · Written 25 Sep 2026
Includes the uncertain-wait loop and quiet review queue

Negotiation: when the other side has done its homework
AI can make negotiation preparation cheaper. The advantage comes from falsifiable questions, visible updates and human-owned authority—not a machine that claims to know the other side.
Subjects: Procurement and vendor selection · Customer decisions and buying journeys · Human judgement and cognitive effects · Evidence literacy and source evaluation
Published 30 Oct 2026 · Written 25 Sep 2026
Includes the Preparation-to-Agreement Loop

The best AI marketplace may be the one that lets you leave
AI agents can widen discovery without removing the marketplace. The stronger platform earns execution through value rather than obstructed exit.
Subjects: Business models and market economics · Agent systems and orchestration · Procurement and vendor selection · Supplier continuity, portability and exit
Published 30 Oct 2026 · Written 18 Sep 2026
Includes the Open Discovery / Preferred Execution Test

Will AI take your job all at once or one task at a time?
A job can keep its title while the economics of its tasks change. Separate capability, adoption, task economics and workforce effects, then map your own recurring work.
Subjects: Workforce and employment · Workflow transformation · Model capability and reliability
Published 30 Oct 2026 · Written 24 Sep 2026
Includes the Task Tide Map

The price of an AI agent is not its token price
Token price is an engineering input, not a procurement outcome. Compare agent offers by the cost of accepted tasks, including tools, retries, review, exceptions and commitments.
Subjects: AI pricing and operating costs · Procurement and vendor selection · Agent systems and orchestration
Published 30 Oct 2026 · Written 24 Sep 2026
Includes the Outcome-Normalised Agent Cost Test

Your employees may be asking AI about you
Employees may ask AI to interpret a manager or organisation before seeking clarification. Leaders should protect private reflection, limit sensitive disclosure and make the human return path easier to use.
Subjects: Leadership and role design · Human judgement and cognitive effects · Privacy and information governance
Published 30 Oct 2026 · Written 24 Sep 2026
Includes the AI First-Contact Audit

The open model ecosystem is getting an owner
An open platform can preserve formal choice while its defaults, optimisation, sequencing or economics begin to favour its owner. The answer is observable neutrality.
Subjects: Business models and market economics · Procurement and vendor selection · Supplier continuity, portability and exit · System lifecycle and vendor change
Published 30 Oct 2026 · Written 24 Sep 2026
Includes the platform-neutrality test

The AI chasm has moved inside the organisation
Access, individual use and operating-model value are different conditions. Find the level at which adoption has actually stopped.
Subjects: Adoption and readiness · Workflow transformation · Investment cases and value realisation
Published 30 Oct 2026 · Written 21 Sep 2026
Includes the five-level AI Adoption Stack

AI capability changes gradually. Management decisions do not
Capability, adoption and employment run on different clocks. Use six thresholds to prepare for a sudden workflow, supplier or workforce decision.
Subjects: Model capability and reliability · Workforce and employment · AI strategy and portfolio · Adoption and readiness
Published 30 Oct 2026 · Written 22 Sep 2026
Includes the AI Decision Threshold Map

How a negotiation tool could become a marketplace
A buyer-side tool can acquire the ingredients of a marketplace one transaction at a time. Test portability before useful dependence becomes captivity.
Subjects: Business models and market economics · Procurement and vendor selection · Supplier continuity, portability and exit · Customer decisions and buying journeys
Published 30 Oct 2026 · Written 21 Sep 2026
Includes the Negotiation System Portability Test

Your AI vendor changed. Did your risk assessment?
Approval belongs to a defined risk case. Use material-change triggers to decide whether a vendor update should be recorded, reassessed or suspended.
Subjects: System lifecycle and vendor change · AI risk and control evaluation · Procurement and vendor selection
Published 30 Oct 2026 · Written 18 Sep 2026
Includes the AI-vendor material-change trigger list

How much is certainty worth?
Price the specific change that better information, reliability, protection or flexibility makes to a decision—not the feeling of reassurance.
Subjects: AI pricing and operating costs · Procurement and vendor selection · Supplier continuity, portability and exit
Published 30 Oct 2026 · Written 19 Sep 2026
Includes the Certainty Premium Test

Not all cognitive friction is waste: what should AI automate, and what must humans still practise?
A three-minute workflow can save time while weakening judgement or moving verification work elsewhere. Separate waste from useful practice, vigilance and new AI-created burden.
Subjects: Human judgement and cognitive effects · Workflow transformation · Accountable decision-making
Published 30 Oct 2026 · Written 17 Sep 2026
Includes the Cognitive Load Allocation Audit

Your human in the loop may be a liability sink
A named reviewer is not automatically a safeguard. Test whether the person can see enough, act safely and carry only the responsibility the workflow lets them exercise.
Subjects: Human oversight and escalation · Accountable decision-making · AI risk and control evaluation
Published 30 Oct 2026 · Written 23 Oct 2026
Includes the Human Oversight Capacity Test

Your procurement agent may know the market. Does it know your deal?
A large pricing dataset supplies a market prior, not deal truth. Authority should depend on evidence for the transaction in front of the buyer.
Subjects: Procurement and vendor selection · Evidence literacy and source evaluation · Model capability and reliability
Published 29 Oct 2026 · Written 17 Sep 2026
Includes the Relevant Intelligence Coverage test

Your content funnel worked. The buyer still had the same question
A form fill records an exchange. The next asset still has to answer the question that made the buyer act.
Subjects: Marketing and commercial operations · Customer decisions and buying journeys
Published 29 Oct 2026 · Written 16 Sep 2026
Includes the continuity-of-intent test

Your company has an AI-readable shadow strategy
AI can combine ordinary public disclosures into a plausible strategic hypothesis that no single disclosure reveals.
Subjects: Privacy and information governance · Customer decisions and buying journeys · Marketing and commercial operations
Published 29 Oct 2026 · Written 15 Sep 2026
Includes the Public Information Aggregation Audit

Crossing the AI chasm, one business function at a time
AI becomes an organisational capability function by function—not when access spreads, but when work reorganises around a complete enough system to depend on.
Subjects: Adoption and readiness · Workflow transformation · AI strategy and portfolio
Published 29 Oct 2026 · Approved 23 Sep 2026
Includes the Functional Chasm Test

MCP explained: How AI connects to your tools and data
MCP standardises how AI reaches tools and data. Your organisation still governs the authority that travels through the connection.
Subjects: Integration and interoperability · Agent systems and orchestration · AI risk and control evaluation
Published 29 Oct 2026 · Written 14 Sep 2026
Includes the MCP Connection Approval Test

What Anthropic’s AI employment numbers may be hiding
An economy-wide AI scenario can conceal much larger change inside organisations adopting at very different speeds.
Subjects: Workforce and employment · Leadership and role design · Organisational capacity and productivity · Evidence literacy and source evaluation
Published 29 Oct 2026 · Written 13 Sep 2026
Includes the AI-native Workforce Test

Your senior team of 2035 is being hired now
Today’s smaller entry cohorts can return later as scarce experience, higher wages and a thinner leadership pipeline.
Subjects: Workforce and employment · Skills and training · Leadership and role design
Published 29 Oct 2026 · Written 13 Sep 2026
Includes the Workforce Cohort Test

When should you upgrade your AI?
Upgrade when a new model improves the operating system—not simply because the model is newer.
Subjects: System lifecycle and vendor change · Model capability and reliability · AI risk and control evaluation · AI pricing and operating costs
Published 29 Oct 2026 · Written 13 Sep 2026
Includes the six-gate AI Upgrade Gate

Can the EU AI Act apply to a business outside the EU?
Test direct supply, returning output and linked value chains before treating geography as a safe harbour.
Subjects: Regulation and compliance · Procurement and vendor selection
Published 29 Oct 2026 · Written 2 Aug 2026
Includes the ten-question EU reach test

Evals are the missing management layer
Turn judgement, risk and organisational intent into repeatable evidence and decisions.
Subjects: AI risk and control evaluation · Governance design and assurance · Accountable decision-making
Published 29 Oct 2026 · Written 2 Aug 2026
Includes the Management Eval Canvas

Beyond the Magic Quadrant
Choose the axes before scoring—and give every quadrant a different action.
Subjects: AI strategy and portfolio · Investment cases and value realisation · Evidence literacy and source evaluation
Published 29 Oct 2026 · Written 3 Aug 2026
Includes the Two-Axis Decision Map

AI value is multiplicative
One weak condition can suppress the return on the whole investment.
Subjects: Investment cases and value realisation · AI strategy and portfolio · Adoption and readiness
Published 29 Oct 2026 · Written 3 Aug 2026
Includes the Limiting-Condition Decision Toolkit

The AI future arrived sideways
The categories were often right. The route was wrong.
Subjects: Model capability and reliability · Workforce and employment · Evidence literacy and source evaluation
Published 29 Oct 2026 · Written 5 Aug 2026
Includes the AI Future-Assumption Audit

Chief AI Officer: permanent role or transitional fix?
Centralise coherence while distributing accountability—and review the title before it persists by inertia.
Subjects: Leadership and role design · Accountable decision-making · Adoption and readiness
Published 29 Oct 2026 · Written 10 Aug 2026
Includes the CAIO Mandate Test

AI project failure may get worse before it gets better
Improve the odds through harder selection, production-level tests and explicit kill criteria.
Subjects: AI strategy and portfolio · Adoption and readiness · Investment cases and value realisation
Published 29 Oct 2026 · Written 11 Aug 2026
Includes the AI Project Kill Test

Your AI strategy may survive the bubble. Will your vendors?
Design the capability to outlive today’s suppliers, platforms and pricing models.
Subjects: Supplier continuity, portability and exit · Procurement and vendor selection · System lifecycle and vendor change
Published 29 Oct 2026 · Written 5 Aug 2026
Includes the AI Capability Survival Test

The AI readiness gap is a leadership gap
Why AI access and activity do not automatically become dependable organisational value.
Subjects: Adoption and readiness · Leadership and role design · Workflow transformation · Investment cases and value realisation
Published 28 Oct 2026 · Written 2 Aug 2026
Includes the AI readiness gap diagnostic

AI is no longer just a tool. It is a leadership capability
Leaders must frame its purpose, redesign work, set standards and remain accountable for the result.
Subjects: Leadership and role design · Workflow transformation · Accountable decision-making · Adoption and readiness
Published 28 Oct 2026 · Written 2 Aug 2026
Includes the AI leadership capability test

Hidden AI in supplier delivery
You may not be buying AI. Your suppliers may be bringing it anyway.
Subjects: Procurement and vendor selection · AI risk and control evaluation · Regulation and compliance
Published 28 Oct 2026 · Written 29 Jul 2026
Includes the supplier AI disclosure and control test

The AI provenance trap
When “Who wrote this?” replaces “Is this any good?”
Subjects: Audit trails and provenance · Publishing and knowledge integrity · Evidence literacy and source evaluation · Human judgement and cognitive effects
Published 26 Oct 2026 · Written 16 Aug 2026
Includes the three-gate AI work review
No analysis found
Try a broader search or another filter.
More analysis will appear here as it is published
The library will remain deliberately honest about what exists—no inflated counts and no placeholder articles.