10pmAI
It's 10pm.Do you know what your AI is doing?

Personas

← Personas
Lagertha persona

Lagertha

CTO

"Corporate sent the AI strategy and the SVP said execute the strategy. I have three questions and no answers. What is the strategy, in terms my engineers can build? Who in my unit is already doing what with AI, with which data? And what exactly am I accountable for when it goes wrong? Six pilots nobody approved, a consultant who is certain, and my engineering lead asking every morning where the data goes. Everyone is sure. Nobody can show me."

Who

CTO of a business unit, about 500 people. Reports to the SVP who said "execute the strategy." Works with his engineering lead, who keeps asking where the data goes.

Elevator statement

As Lagertha, I want to know what the AI strategy means for my unit, who is already doing what with which data, and what I am accountable for, so that I can turn the order into a plan that rests on facts and stands up next to the other units, because I sign it and "the consultant said so" is not a defense.

Solutions

CTO AI Ground Truth Assessment

Today

One business unit, about 500 people. Corporate systems with AI features that arrived in updates. Engineering wired an LLM into the product. Marketing runs a pilot on a company card. Three vendors offer governance platforms with free trials. No inventory, no data-flow picture, no owner.

What is known

He has been told exactly what AI is for, by everyone, and still cannot see it. The strategy is not his but the outcome is. He will not install a platform, or a Guardian, before he knows what it is supposed to govern.

Buy trigger

The SVP sets a review date. A peer unit's pilot blows up and every CTO gets the same question. A vendor's trial hands him a report he cannot interpret. A customer or auditor asks what AI touches their data and three people give three answers. Or his engineering lead was right.

Around him

Ragnar, the SVP he reports to, who wants one page across all units. His engineering lead, who is right. Peer CTOs with the same order. The consultant or vendor who is certain and wrong. Would not buy: a Guardian, a platform before an inventory, another roadmap.

Solution · Service, not a Guardian

CTO AI Ground Truth Assessment

Three weeks, fixed fee, one unit. AI portfolio map, data-flow map, evidence-graded forecast, three moves. All three questions answered in one document.

“I give unit CTOs ground truth on their AI: what is real, what is stuck, where the data goes. Three weeks, fixed fee, three moves.”

What this solution needs to deliver

  • A unit-level map of reported versus actual status for every initiative, slides or not
  • Per-initiative data-flow trace counsel can read: what leaves, to whom, under what terms, who approved
  • Findings graded by their own evidence (E1 to E4): verified versus somebody's word
  • Value-stream method from someone who shipped the hardware and firmware
  • Independence that is structural: no remediation credit, no platform to sell

Problem evidence

Independent sources cited in the one-pager, October 2026

Figures reproduced from the October 26, 2026 persona one-pager and attributed there to the sources below; not independently verified.

80%
of AI initiatives fail to deliver intended value; 34% abandoned before productionRAND, 2,400+ initiatives
95%
of GenAI pilots show no measurable P&L impactMIT NANDA, 300+ deployments
42%
of companies abandoned at least one AI initiative in 2025, up from 17%S&P Global
61%
of projects approved on projected ROI never measured after launchMIT Sloan 2025
43%
of organizations cannot audit or inventory the AI tools in use; only 34% detect shadow AIGartner 2025
60%
of security teams lack visibility into which AI tools employees useCisco AI Readiness Index
21%
have a mature AI governance framework (Deloitte 2026); 52% have no external AI tool policy (KPMG)
77%
of AI failures are organizational, not technical140 implementations, 2026

Takeaway: A map beats a framework, and the unit CTO who has one stands up better next to the four who do not.

Sources cited in the one-pager: RAND 2025, MIT NANDA 2025, S&P Global 2025, Gartner 2024 to 2026, Cisco, KPMG, Deloitte 2026, MIT Sloan 2025; vendor pages EPC Group, Veracity AI, 10Alyx, Aries Consulting, Christian & Timbers guide.

Who sells something adjacent

5 adjacent options · October 2026 research notes

Offerings, prices and gaps are the one-pager’s assessment, not independently verified. Vendor claims and pricing may change.

EPC Group
"AI Deployment Second Opinion": review of one engagement, board memo, three corrections
$15K fixed, 10 days; fee credited to their remediation
Gap: One project, Microsoft only, independence asserted not structural
Veracity AI
Governance readiness assessment, TRACE score, gap report
$8K to 12K, 2 to 3 weeks
Gap: Policy maturity. No initiative map, no data-flow trace
Aries Consulting
AI readiness audit, 90-day roadmap
$7,500, 2 to 3 weeks
Gap: SMB founders. Readiness to start, not truth about what runs
Big Four, McKinsey, Accenture, IBM, Protiviti
Maturity diagnostics, EU AI Act mapping, AI assurance
Unpriced, inside larger advisory
Gap: Slow, heavy, conflicted with implementation
Governance platforms
Credo, Holistic, Trustible, Airia: inventory and monitoring software, free trials on his desk
$50K to 200K per year
Gap: Tooling someone still has to populate. Channel, not competitor