Edition in partnership with

Welcome executives and professionals. Token economics is a material risk and opportunity that continues to draw attention across the C-suite and board.

Since the previous edition, we have reviewed hundreds of the latest insights in agentic and generative AI, spanning best practices, case studies, market dynamics, and innovations.

This briefing outlines what is driving material value — and why it’s important.

In today’s briefing:

• The CIO's guide to AI tokenomics.

• Enterprise AI factory trends.

• The AI First organization.

• PwC's AI-powered strategy solution.

• Transformation and technology.

• Insights for Executive+ members.

• Career opportunities & events.

Read time: 4 minutes.

MARKET & BEST PRACTICE INSIGHT

The CIO's guide to AI tokenomics

Brief: Accenture's 41-page AI tokenomics guide for CIOs details how CIOs can see, control and account for AI token spend at scale, drawing on a survey of 750 senior executives at enterprises with revenue above $1 billion.

Breakdown:

• Under one dollar in five of enterprise token spend can be tied to a quantified financial outcome that finance is able to act on.

• Enterprises expect token consumption to grow 78% over the next 24 months, even after a predicted 19% decline in the price per token.

• Firms pulling ahead build observability into workloads before production, and bill teams for the tokens they consume via chargeback.

• They route each task to the lowest-cost capable model, demand a measurable value case before launch, and train staff to use AI efficiently.

Why it’s important: Boards are pressing CEOs to explain the exposure and return, CEOs are looking to CFOs for control, and CFOs are turning to CIOs to make AI consumption predictable and valuable. Those that master these economics will gain an advantage over those that simply pay the bill.

IN PARTNERSHIP WITH RESOLVE

Make AI agents actionable, governable, and worth It

Brief: Resolve AgentLab turns operational expertise into governed AI agents using natural language. Describe the work, define the context, and connect agents to deterministic automation so they act reliably across your systems.

How is Resolve different?

• AI reasoning for context, deterministic workflows for predictable execution

• Build agents and automation in natural language, no developers required

• Turn operational expertise into reusable, governed agent skills

• Orchestrate actions across departments, and systems without lock-in

See AgentLab for yourself on September 16.

MARKET & BEST PRACTICE INSIGHT

Enterprise AI factory trends

Brief: Deloitte explored AI factory trends, covering the motivations driving adoption, how quickly adoption is projected to accelerate across five industries, and the payoff organizations can expect from building them.

Breakdown:

• Motivation for building AI factories can vary widely by industry, signaling different competitive, regulatory and operational aims (image above).

• Energy, resources and industrials lead adoption at 50% to 82% by 2028; financial services jumps 24% to 63%, with TMT near 72-75%.

• TMT expects the most token growth, with above 10 billion monthly tokens rising from 29% today to 71% by 2028; FSI follows, 36% to 70%.

• Respondents named innovation capacity (71%), risk management (64%) and token optimization (59%) as the top expected AI factory outcomes.

Why it's important: AI factories help turn pilots into governed enterprise capabilities, aligning infrastructure, data and governance to manage token costs, data latency and sovereignty needs. AI factories can also enable firms to sell GPU as a service, or their own solutions externally.

BEST PRACTICE INSIGHT

The AI First organization

Brief: Arthur D. Little's viewpoint explored what it takes to become an AI First organization, focusing on how the IT function can help reshape the wider enterprise operating model so that AI delivers on its full potential.

Breakdown:

• Early "human-AI sandwich" models leave existing workflows unchanged, delivering 10% to 20% gains where a full redesign could reach 50%.

• Becoming AI-first means designing the entire operating model as if AI capabilities had existed within the business from day one.

• Transforming IT to enable the shift requires attention to building blocks spanning architecture, operating model and human factors (image above).

• The viewpoint details seven actions for making the transition, alongside four emerging design principles drawn from first movers.

Why it’s important: Integration of AI agents into enterprises is gathering pace, but most initial moves target quick-win productivity gains that deliver only a fraction of AI's potential. Redesigning the operating model around humans and AI together is the ultimate goal.

AI-NATIVE PROFESSIONAL

How to onboard new hires with ChatGPT

Brief: In this guide, you'll learn how to use ChatGPT to collect approved new-hire context, build tracker updates, draft team-by-team summaries, and plan the welcome-space setup, all assembled for review before anything is sent.

Step-by-step:

1. New-hire onboarding spans several systems: an accepted-hire list, a tracker, team mappings, equipment readiness, and chat spaces.

2. ChatGPT can help coordinate it: use the starter and follow-up prompts to inventory a start-date cohort and prepare tracker updates.

3. Then ask it to summarize the batch team by team and draft the welcome-space setup, all gathered into a single reviewable pack.

4. Always keep the first pass read-only, then explicitly approve any tracker writes, invites, posts, DMs, emails, or channel creation.

Best practice: Onboarding data is sensitive: never include compensation, government IDs, home addresses, medical details, or performance notes.

For the full guide, including prompts, upgrade to Executive+ or The Boardroom.

ACCELERATOR

PwC's AI-powered strategy solution

Brief: PwC launched an AI-powered strategy solution that compresses months of early-stage strategy work into days without losing rigor, so senior effort shifts toward refining recommendations and securing organizational buy-in.

Breakdown:

• It is built on PwC gold data such as sector benchmarks, its Customer Link platform, client context and a shared semantic memory.

• It works across three pillars: the what, meaning problem statement and pathways, the how, or transformation required, and execution.

• An orchestrator routes tasks to retrieval and analysis agents, then a reflection agent checks evidence before finalization (image above).

• Work that once took six to eight weeks of senior-led effort now reaches a first recommendation in days, with every step traceable.

Why it’s important: Compressing the early-stage analysis frees senior time for the harder part: refining the recommendation, contextualizing it and building the stakeholder alignment that turns a strong answer into a decision the organization will actually back and fund.

IBM published a 24-page report on five steps to capture higher AI ROI, from aligning spend with performance to cutting tech debt.

Deloitte explored four plausible futures for multi-agent enterprise AI and examined the future of human-AI decision-making at work.

Oliver Wyman surveyed 200 executives on agentic AI, introduced agentic clock speed, and explained why you can't buy your way to it.

McKinsey detailed what reinventors do differently to create AI value and met Lloyds CEO Charlie Nunn to discuss AI trust and optimism.

PwC answered questions on creating AI-native workflows with agentic scaffolding, from reinventing work to scaling agent networks.

Microsoft shared how Microsoft AI is helping Cognizant scale delivery excellence across 10,000 projects and thousands of engineers.

OpenAI launched ChatGPT for Financial Services with PitchBook, Crunchbase and LSEG data, and released the Agents API in public beta.

Salesforce introduced a Trusted Enterprise AI Harness with six capabilities and an AI Control Plane to govern agents across the org.

Anthropic published its latest Threat Report, 154 pages detailing cases of Claude misuse it disrupted, from bioweapons to espionage.

OpenAI introduced a Data agent in ChatGPT Work that turns company data into answers, interactive dashboards, and action.

Microsoft detailed the Citadel architecture, building on Azure AI landing zones to turn enterprise AI into a shared operating model.

Anthropic researcher expressed 10%+ odds AI could kill all humans; David Sacks called it a “doomer psyop” for regulatory capture.

CAREER OPPORTUNITIES

AWS - AI Transformation Principal

BNY - AI Strategy Senior Vice President

Visa - AI Transformation Director

EVENTS

Responsible AI Summit - September 21-23, 2026

C3 AI - AI Accelerator - September 23, 2026

C-Vision - AI-Driven Cybersecurity - September 29, 2026

Reach enterprise AI decision-makers:

• 66% of readers are C-level executives or VP and Director-level leaders.

• 63.2% of the audience is based in the U.S., EU, UK, ANZ, and Singapore.

• Read by leaders at Microsoft, Deloitte, the Fortune 500, and more.

Guaranteed impression and custom sponsorship packages available, with post-send performance reporting.

Conceived as a practical communication for executives Lewis Walker has worked with, this briefing has become a trusted resource for thousands of senior decision-makers shaping the future of enterprise AI.

We welcome your feedback.