- Enterprise AI Executive
- Posts
- 24 playbooks for AI executives
24 playbooks for AI executives
Plus, key takeaways to help you level up fast.
Lewis Walker
August 11, 2026
Welcome executives and professionals. AI is a generational opportunity for executives to transform their enterprises and strengthen their market position. But many are at risk of getting burned by spiraling costs, security risks, sovereignty concerns, and board-level misalignment.
The right strategies can help executives overcome these challenges. I reviewed 427 playbooks from the past 10 weeks. These 24 are must-reads for AI leaders:
MCKINSEY & COMPANY
Image source: McKinsey & Company
Brief: McKinsey shared an A-E playbook for agentic change: how leaders can close the adoption gap by taking their employees from awareness to belief to commitment to capability and, finally, to a reinforced operating system.
Breakdown:
Traditional change management is not enough. Periodic comms, and top-down rollouts build awareness, not trust and discipline.
Closing that gap needs a change leadership approach, C4 reinvention, that puts people at the centre of creating value from AI.
Move employees deliberately through five stages: awareness (A), belief (B), commit (C), develop (D), and finally to enforce (E).
Make the change tangible, have leaders adopt first, give people room to choose, build competence in the work itself, then enforce.
Why it’s important: Successful AI transformations follow a 1:3:5 pattern. For every dollar invested in agentic technology, organizations spend three on process redesign and five on capability building and adoption. Change leadership sits with CEOs, CFOs, and CHROs, who each play distinct and complementary parts.
BOSTON CONSULTING GROUP
Image source: Boston Consulting Group
Brief: Boston Consulting Group shared how, as AI becomes central to enterprise decision making, CEOs now face a challenge: avoiding AI vendor lock-in and protecting what makes their business unique.
Breakdown:
Providers across the AI ecosystem, from frontier labs and hyperscalers to open weight developers, are racing to own the stack.
Past waves show how difficult it can be to unwind dependencies once a vendor platform becomes essential to daily business operations.
CEOs need a layered AI stack with a security perimeter around their most valuable knowledge: the enterprise cortex, the firm brain.
That cortex is your IP, essential data, business rules, and codified understanding of how processes link to strategy and values.
Why it’s important: Technological lock-in is becoming cognitive lock-in, where organizations depend not just on a platform but on AI reasoning that shapes how they think and operate. A modular architecture lets you adopt the best AI models as they evolve, preserving autonomy and advantage.
GOOGLE CLOUD
Image source: Google Cloud
Brief: Google Cloud's 66-page AI Transformation Framework provides a repeatable, three-layered approach to help organizations move beyond simple automation toward a cohesive ecosystem of intelligent agents.
Breakdown:
Vision and strategy: Move from AI that answers to AI that acts, tying each initiative to efficiency, cost, customer experience, and revenue.
Agentic development lifecycle: Accelerate time-to-value with an iterative cycle of reimagining, prototyping, building, and embedded quality.
Horizontal and foundational capabilities: Scaling AI hinges on governance, efficient operations, quality data access, and security.
Throughout the framework, Google highlights the key activities, roles and responsibilities, and outputs and deliverables involved.
Why it’s important: Many organizations already run impressive pilots, but the leap to a scalable, secure, enterprise-wide capability is significant. Without a deliberate plan, experiments become disconnected silos with inconsistent governance, technical debt, and rising costs.
MICROSOFT
Image source: Microsoft
Brief: Microsoft published a 44-page playbook on becoming a frontier firm, detailing its evolving methodology for driving AI transformation, built around five elements that together enable AI to deliver lasting business outcomes.
Breakdown:
Anchor AI in service to your business strategy, and build an AI operating model that delivers against your business goals at scale.
Adapt roles, manager behaviors, and learning so that AI expands human capability and productivity, rather than eroding it.
Codify what makes your company distinct into evals and a learning system, then keep that intelligence compounding.
Safeguard your security by setting guardrails across all agents, ensuring every AI action stays traceable, auditable and reversible.
Why it’s important: Transformation is hard: it requires people to fundamentally rethink how they work. Reaching the frontier takes courage and sustained leadership. When leaders give people clear intent and real agency, they rise to meet the challenge and reinvent how work gets done.
DELOITTE-GOOGLE CLOUD
Image source: Deloitte
Brief: Deloitte AI Institute, with Google Cloud, published a 21-page framework for scaling agentic AI that focuses on technical approaches and human architecture while emphasizing real-world applications and strategies.
Breakdown:
The pilot-to-agentic enterprise gap centers on three friction points: technical fragility, process inertia and trust deficits.
Enterprises need a “pragmatic AI” mindset: value over velocity, specialized utility over AGI, outcome-driven roadmaps.
Adopt a holistic framework for a scalable AI architecture: an agentic OS, developer workbench, an agent repository, and more.
Beyond technology, insights span human architecture, from executive sponsorship to a culture of iterative failure.
Why it’s important: In the near term, one to two years out, internal agent marketplaces will be standard in large enterprises. By 2028, cross-enterprise collaboration will emerge, where, for example, a retailer's inventory agent negotiates with a supplier's shipping agent to restock autonomously.
PALANTIR
Image source: Palantir
Brief: Palantir published a 28-page blueprint outlining the 15 steps every company and government must take to protect their sovereignty and compound lasting competitive alpha in this fast-moving age of AI.
Breakdown:
The first three steps cover foundations: ensure zero data retention, map your AI decision tree, and identify architecture opportunities.
Next, the model layer: guard against misaligned incentives, maximize model liquidity, and own the model flywheel.
The compute layer follows: decide hardware based on assurance, own adaptable hardware for sensitive workflows, verify outside compute.
Finally, the control layer covers agnosticism, permissions, auditing, cybersecurity practices, branching, and context flywheels.
Why it’s important: Sovereignty is your alpha. Many enterprises are being led to believe their choices are narrower than they are, leaving critical decisions about how AI is applied in the hands of others. In reality, you have complete agency over your model usage. This is a guide through those decisions.
BAIN & COMPANY
Image source: Bain & Company
Brief: Bain & Company, drawing on its recent survey of 100 CEOs, explored how to win with AI, including the seven CEO decisions that differentiate companies building proprietary intelligence.
Breakdown:
Most CEOs think they're leading an AI transformation, but they're managing a portfolio of pilots, and the two are not the same thing.
The companies pulling ahead are building proprietary intelligence through unique data, encoded workflows, and learning architectures.
Seven CEO decisions differentiate leaders: posture, domain focus, data, tech architecture, operating model, learning, governance.
Three leadership traits binding the seven choices together: personal commitment, ruthless conviction, and investing for learning.
Why it’s important: It is easy to say "yes" to the decisions that create proprietary intelligence. Whether you have made them shows up elsewhere: in calendars, budgets, who sits on which committee, and what reaches the board. None can be delegated; all sit with the CEO.
BOSTON CONSULTING GROUP
Image source: Boston Consulting Group
Brief: BCG shared insight on perhaps the most underappreciated, yet most valuable, use of decision agents: supporting executive committees as they navigate the complex, high-stakes decisions in the boardroom.
Breakdown:
Customized decision agents help senior, cross-functional committees navigate the complex, high-stakes decisions they face daily.
Integrated business planning requires decision agents to align volume, capacity, and budget decisions across functions.
Portfolio investment allocation: agents help synthesize capital commitments that require strategy, finance, and operations input.
Market-entry decisions: agents help assess risk, feasibility, and timing together for faster, better-informed strategic choices.
Why it’s important: A custom-built decision agent acts as another team member in the boardroom: an omniscient chief of staff with supercomputing power that gathers and synthesizes inputs, tests scenarios, and then formulates concrete recommendations and follow-up actions.
ANTHROPIC
Image source: Anthropic
Brief: Anthropic's Deputy CISO, Jason Clinton, shared his team's lessons learned adopting agentic AI, including how every agentic use case that reaches their review process is assessed for risks using four core questions.
Breakdown:
What untrusted content does it ingest? Untrusted means anything an attacker could plausibly write or alter, like outside email or the web.
What actions can it take, and on whose behalf? Tool calls, code execution, and network egress each widen the aperture and the risk.
What is the blast radius if misaligned? Could it access one file or the whole org, would it be an anomaly, an annoyance, or a true breach?
What observability do you have? Can you distinguish agent actions from user actions, and does that activity land in your SIEM logs?
Why it's important: A CISO's responsibility in the age of agentic AI is not to achieve zero risk. Instead, it is to make agentic risk legible and bounded, so organizations can deliberately accept the risks they can manage, allowing the business to move on its own terms.
MCKINSEY & COMPANY
Image source: McKinsey & Company
Brief: McKinsey, drawing on a survey of nearly 4,000 buyers and sellers across 13 countries, outlined five priority impact journeys that can be rewired with agentic AI to improve the sales lifecycle and drive growth.
Breakdown:
Chapter 1 lays out five priority impact journeys that place agentic AI at the center of how commercial teams sell and grow.
They span finding opportunities, the right go-to-market model, the right offer, and the right price, every single time (see image).
Chapter 2 details the operating model shifts needed to make the change stick, scale adoption, and turn gains into margin improvement.
The insights are informed by interviews with enterprise executives and case studies from companies achieving early success.
Why it’s important: This is not a more automated version of transactional selling but an insight-led model of solution selling that frees humans to deepen relationships and improve outcomes. At its core is a shift from fragmented use cases to end-to-end impact.
BOSTON CONSULTING GROUP
Image source: Boston Consulting Group
Brief: BCG outlined five moves CEOs should take to close the AI knowledge gap with their board directors. With more than 60% of CEOs saying their boards are rushing AI transformations, alignment is a leadership imperative.
Breakdown:
CEOs know where the company stands, so they should personally frame what AI means for value creation, operations, and risk.
Rather than delegating to technology leaders, CEOs can take a more active role in educating directors through hands-on sessions.
Visits to AI-first companies, customer conversations, and competitor scans give directors firsthand perspective on the wider market.
Boards often push to replace workers faster than is wise, so CEOs must show them where human judgment still adds the most value.
A dedicated transformation committee drawn from directors with stronger AI fluency can close the gap faster than the full board.
Why it’s important: CEOs often carry more of the AI decision-making burden than they or their boards think they should. That gap means CEOs must do more than bring boards up to speed on AI; they must also set expectations for how each senior leader drives value creation.
PALANTIR
Image source: Palantir
Brief: Palantir highlighted the gap between agentic performance, safety, and reliability in theoretical versus enterprise settings, and how firms can understand and improve results if agents fall short of expectations.
Breakdown:
Robust governance infrastructure for evaluating enterprise AI agents is essential to ensuring agent performance, safety, and reliability.
Palantir sorts the governance capabilities that should sit at the core of the AI system into two layers: controls and workflows.
Governance controls form the foundation: authorization workflows, bounded execution, testing, evaluation, observability, and fail-safe modes.
Governance workflows build on those mechanisms so users can realize their aims: human-agentic collaboration and lifecycle development.
Why it’s important: Palantir draws these insights from deploying its software across enterprises and institutions for mission-critical outcomes. Built atop existing governance processes, the practices help guide how leaders and policymakers approach AI oversight and regulation.
OLIVER WYMAN
Image source: Oliver Wyman
Brief: Oliver Wyman, with proSapient, surveyed 100 sales leaders at companies already using agentic AI, revealing where it delivers the most impact and where scaling requires new capabilities and operating models.
Breakdown:
Sales leaders most often cite a positive impact of agentic AI on sales growth (89%), productivity (87%), and lead conversion (61%).
Sales leaders overwhelmingly report strong agentic AI impact at the top of the funnel, where reps find and prioritize prospects (image above).
Fewer sales leaders saw meaningful mid-funnel impact, where leads are nurtured and handoffs occur between AI agents and human reps.
37% of sales leaders said enhanced decision-making was the single most effective change agentic AI has brought to their sales organizations.
Why it's important: The next wave of value will be captured by leaders who pilot and scale agents systematically, with integrated data, clear governance, and well-choreographed human/AI handoffs that protect customer experience. Acting now with focus will build lasting advantage.
BOSTON CONSULTING GROUP
Image source: Boston Consulting Group
Brief: BCG published the second in a series of insights on the cost of AI tokens and how companies can manage them. The first article examined the true costs of AI; this one details the challenges of measuring them in practice.
Breakdown:
AI works differently than traditional software as a service, changing the unit of management. Traditional FinOps isn't built for it.
Instead, a ratio BCG calls return on AI (RoAI) captures the full cost of the AI being applied across the business and its outcomes.
To account for costs and assess RoAI, companies need a workflow-level operating model that enables management to do three things well.
See what's happening, shape the cost, and either prove the value or stop (or minimize) the activity, as detailed in the image above.
Why it’s important: As AI moves into production, CFOs, CIOs, and CTOs inherit a substantial new cost: tokens consumed to produce outcomes. Rising agent use pushes the meter into overdrive. FinOps cannot keep pace, and CEOs will expect the C-suite to rise to the challenge.
WORLD ECONOMIC FORUM
Image source: World Economic Forum
Brief: The World Economic Forum, with Kearney and drawing on insights from more than 50 leading organizations, published a 52-page report on how AI-first enterprises are rethinking business models, workflows, and decision-making.
Breakdown:
AI-first success rests on five blocks: intelligence engines, adaptive stacks, redesigned operations, human-AI teaming, new value.
AI-first and AI-native firms are showing speed, scale, and leverage, but which model proves dominant over time is not yet clear.
For incumbents, the question is not whether to become AI-first but how fast and where to start. The window is open, but not forever.
Case studies from Indeed, Gamma, and Cognizant show how firms embed intelligence at scale to unlock innovation and productivity.
Why it’s important: The challenge is not to assume a single operating model, but to build capacity to learn and adapt. The report recommends running parallel models, AI-first alongside existing, and measuring the difference to find where intelligence creates real advantage.
OPENAI
Image source: OpenAI
Brief: OpenAI outlined practical steps for enterprise leaders to understand how AI is used across their organizations, control spend through targeted policies, and direct investment towards work that creates the most value.
Breakdown:
Leaders need a plain view of AI usage: who uses it, which models, how much capacity, and what kind of work it supports.
A more capable model may cost more per token, yet it can reach an acceptable result faster, with fewer attempts and less review.
Spend controls like workspace defaults, group limits, and overrides let leaders support high-value work without raising limits broadly.
OpenAI also addresses managing AI investments as a portfolio and matching the product, capacity, and support model to workflow demand.
Why it’s important: Token price alone does not show whether AI creates value. Leaders should look at useful work per dollar: tasks completed, time saved, and decisions improved. As teams move from chat to longer-running workflows, enterprises need clearer visibility into demand, spend, and risk.
INFOSYS
Image source: Infosys
Brief: Infosys published a paper giving executives a plain-language framework for understanding enterprise AI token economics: what drives costs, how to forecast them, how to govern them, and what the ROI of doing so looks like.
Breakdown:
AI token costs are poorly governed not because they're uncontrollable, but because their governance frameworks are less understood.
Build observability first, you cannot govern what you cannot see. Then add AI model routing: lowest effort, highest return.
Deploy quality guardrails in audit mode first, then enforce; add workflow budget controls as automated pipelines mature.
Executive ownership of governance decisions (image above) is consistently the single biggest predictor of a program's cost outcomes.
Why it’s important: AI token costs are architecture-driven, not user-driven: two decisions, how you route AI model calls and how you govern information retrieval, drive 78% of achievable savings. Enterprises that build governance before they scale spend 40-50% less, with no loss of capability.
GENPACT
Image source: Genpact
Brief: Genpact and HFS Research surveyed 2,000+ enterprise executives across 16 industries and 14 functions, identifying nearly $18 trillion in recoverable value and four interconnected enterprise debts holding it back.
Breakdown:
Data debt is the gap between data firms have and what AI needs, only 33% is AI-ready and 42% of projects fail due to data quality issues.
Process debt is the cost of manual, ungoverned workflows that waste 40% of weekly employee time. AI just executes the wrong steps faster.
Tech debt is the legacy infrastructure tax: core enterprise systems average 10 years old, and 42% of developer time is spent servicing them.
Talent debt is the workforce's readiness gap for AI: only 32% are AI-ready, and it silently amplifies every other debt.
Why it’s important: Resolving these debts (image above) unlocks 8% faster revenue growth and 16% lower costs, yet 85% say they actively limit AI value and over half lack a funded plan to address them. With ~13% of function spend now on AI, the ambition-foundation gap has never been costlier.
ACCENTURE-CMU
Image source: Carnegie Mellon University
Brief: Carnegie Mellon University (CMU), with Accenture, built The AI Adoption Maturity Model, a practical 63-page framework to help leaders assess where they stand, which capabilities to strengthen, and how to scale AI confidently.
Breakdown:
The model defines five maturity levels of AI adoption: exploratory, implemented, aligned, scaled, and future-ready.
Maturity is assessed across four organizational change dimensions: strategy, workforce, workflow re-engineering, and risk.
It also spans four AI lifecycle engineering dimensions: data, engineering, operations, and the broader technology ecosystem.
Each capability area sets out its relevant goals, practices, and artifacts to guide maturity and consistent execution.
Why it’s important: The model has been deployed at several Fortune 500 companies through an early-adopter program, where it delivered strong results and proved effective at accelerating disciplined, enterprise-scale AI adoption. Results are now cited verbatim in boardrooms.
Images source: Google
Brief: Google published a 51-page whitepaper on the new SDLC built with vibe coding and agentic engineering, as AI compresses implementation while requirements, architecture and verification stay stubbornly human-paced.
Breakdown:
It maps the spectrum from vibe coding to agentic engineering, conductor-to-orchestrator roles, and factory model of software production.
Structure scales, vibes don't. Vibes suit prototyping, but software that firms rely on needs agentic engineering specs, tests, and oversight.
AI amplifies engineering culture. Strong testing, clear standards, and reviews yield more value; AI multiplies strengths and flaws.
The human role is evolving, not diminishing. Skills are shifting from implementation to judgment, from writing code to designing it.
Why it’s important: The shift from syntax to intent is a present reality. Developers are already spending more time describing what they want than how to build it, and the SDLC is being reshaped around AI. The question isn't whether this happens, but how well enterprises navigate it.
BOSTON CONSULTING GROUP
Image source: Boston Consulting Group
Brief: BCG's 24-slide executive playbook, Driving Sustained Structural Cost Advantage with Applied AI, details how AI leaders achieve three times greater cost reduction than laggards, building lasting structural cost advantage.
Breakdown:
Most AI cost-cutting programs are failing to deliver: copilots layered onto existing workflows fail to move the needle on costs.
AI leaders drive deeper integration, pairing it with traditional cost-cutting strategies, and managing both savings and costs it creates.
Companies fall into common traps: fragmented initiatives, with AI added on top of processes instead of eliminating or consolidating them.
Other traps stall momentum: weak early proof points delay investment, and targets set on productivity but not P&L outcomes.
Why it’s important: Avoiding these traps requires CEOs and CFOs to take direct ownership of the AI cost agenda, demanding a fundamental rethink of operating models and pushing AI strategies that deliver measurable, sustained cost-saving impact across the enterprise.
MCKINSEY & COMPANY
Image source: McKinsey & Company
Brief: McKinsey outlined the data challenges that hold companies back from scaling AI, and set out the practical steps Chief Data Officers (CDOs) can take to overcome them and turn their data into a solid foundation for enterprise AI.
Breakdown:
More than two-thirds of high-performing companies say that data is the primary challenge in capturing value from gen AI (image above).
CDOs should bring structured and unstructured data into data products and establish shared foundation services (e.g. retrieval layers).
Enable federated delivery on top of common infrastructure, and manage derived artifacts like embeddings as core enterprise assets.
Govern semantic consistency across access paths/modalities, and assess readiness against four metrics: reuse, reliability, governance, scale.
Why it’s important: To scale AI across the organization, CDOs must treat data as a core enterprise asset, with the standards and controls that make it reliable, traceable, and usable across applications. Those who manage this balance scale AI safely, consistently, and fast.
PWC
Image source: PwC
Brief: PwC's 2026 Global AI Jobs Barometer analyzed over a billion job ads from six continents, revealing AI is creating a two-track labour market in which skills like judgement and leadership are even more critical and rewarded.
Breakdown:
Since AI use soared in 2022, firms in the most AI-exposed sectors have tripled their productivity lead over the least exposed.
Rather than replacing jobs at scale, leading organisations are using AI to amplify human performance and increase wages.
AI-exposed junior roles are 7x more likely than less-exposed peers to demand traditionally senior skills like strategic thinking.
AI is having two different impacts on jobs depending on whether it is automating more or less expert tasks (image above).
Why it’s important: AI is driving substantial productivity gains for companies and, perhaps surprisingly for some, those seeing the biggest gains are increasing wages and headcount faster than their less AI-exposed peers. AI adoption is expanding opportunity rather than eliminating it.
UBER
Image source: Uber
Brief: Uber's CTO, Praveen Neppalli, shared how agentic AI adoption at Uber is changing the way the company builds, not only in engineering, but across every function of the organization, spanning Finance, Operations, and beyond.
Breakdown:
Today, 99% of Uber's engineers use AI tools, over 70% of pull requests come from agents, and teams have built 2,500+ agent skills.
To extend agentic AI beyond engineering, Uber created Agentic Pods, pairing AI-proficient engineers with business domain experts.
Each pod had just two weeks to go from shadowing an expert to building and shipping a working agent (image above).
In two months, 16 Agentic Pods ran across 16 functions, including cutting capital allocation from 15 hours to 30 minutes.
Why it’s important: The productivity gains are impressive, but what surprised Uber most was how quickly engineers in unfamiliar domains uncovered opportunities hiding in plain sight, by rethinking entire workflows around AI rather than simply automating one isolated task at a time.
MORE MUST-READ BREAKDOWNS
Enterprise AI for CEOs (16 playbooks)
AI for Board Directors (19 playbooks)
Enterprise AI for CXOs (32 playbooks)
AI for Enterprise Leaders (18 playbooks)
Build-Buy-Borrow (14 frameworks)
Agentic AI Case Studies (40 cases)
AI Use Case Prioritization (12 frameworks)
AI Strategy Playbooks (16 playbooks)
AI Agents & Agentic AI Use Cases (2,195)
ENTERPRISE AI EXECUTIVE
Agentic and generative AI are evolving rapidly in the enterprise, driving a new era of AI transformation.
Twice a week, we review hundreds of the latest agentic and generative AI best practices, case studies, market dynamics and innovations to bring you what is driving material value — and why it’s important.
Example editions:
Found this valuable? Share with a colleague.
Received this from someone else? Sign up here.
Connect on LinkedIn.
Lewis Walker, Editor