The CAIO's First 100 Days: An Economic Governance Playbook
A practical guide for newly appointed Chief AI Officers and VPs of AI on building visibility, proof standards, and cross-functional governance in the first hundred days.
·7 min read
CAIOgovernanceAI economics
On this page
Key takeaways
The job beneath the title
Some organisations already have active AI use cases. Others mostly have enthusiasm, vendor conversations, and pilot plans.
A CAIO does not need to know everything before acting, but they do need to make the organisation more capable of answering three questions: what are we spending, who owns the outcome, and what proof standard governs scale?
Real-world pattern: In 2026, Marc Benioff said Salesforce expected to spend about $300 million on Anthropic tokens in 2026, largely for coding-related work. This pattern—where the CAIO's first task is simply establishing what is already happening—is common across enterprises.
Days 1 to 30: build the first view of reality
Build an AI investment register. List current and planned initiatives. Record sponsor, owner, workflow, model or tool path, current stage, spend surface, and intended benefit.
Days 31 to 60: define proof and ownership
For each material initiative, define:
the primary return dimension
the baseline
the named value owner
the expected timeline to proof
the next stage-gate threshold
Proof standard in practice: The UK government's Microsoft 365 Copilot evaluations demonstrate the standard CAIOs should establish: task-level measurement, control groups where possible, quality scored, findings published even when unflattering. This rigour—applied consistently across all major AI initiatives—is what distinguishes governance from theatre.
The CFO needs to know how AI value claims will be evaluated. The CIO needs to know what shared platform and architecture assumptions are being made. Product and engineering leaders need to know what behaviour will be expected before a case is allowed to scale.
Days 61 to 100: establish the operating cadence
FinOps for live demand and unit-cost visibility
TBM for service and capability translation
ITFM for planning and reporting discipline
SPM for portfolio sequencing and funding logic
Engineering and product for workflow design and adoption
Risk, security, and legal for control burden
Discipline integration example: The FinOps Foundation's 2025 guidance on AI scopes emphasises that AI cost management must extend beyond infrastructure to include model and inference costs, data and pipeline costs, and governance costs. A CAIO who positions FinOps as only an infrastructure concern will miss the majority of AI economics.
Methodology note on CAIO prevalence: IBM's 2025 CAIO report found that 26% of more than 2,300 contacted organisations had a CAIO. IBM's 2026 CEO Study reports 76% with a CAIO or equivalent. These figures should not be treated as a simple growth trend. The studies surveyed different populations, used different definitions (especially "or equivalent"), and may reflect rapid role creation, relabelling, or sampling differences. The useful insight is that AI accountability is moving higher in organisations, but title growth does not automatically mean clearer authority or better value governance.
The relationship with CIO, CFO, and CDO
Optimist
Sceptic
The Optimist's Case
The Sceptic's Case
What a good first hundred days produces
1. AI portfolio ledger
A structured register capturing for each initiative:
use case and workflow
named business and technology owners
current stage (explore, prove, scale, operate, retire)
spend surface and run cost
value type and expected benefit
risk tier and control requirements
evidence quality and baseline confidence
next decision date and threshold
2. C-suite decision-rights map
Explicit accountability across CEO, CAIO, CIO/CTO, CFO, CDO, business owners, risk, legal, and CHRO for:
