What are the key takeaways from this HBR IdeaCast episode?

Key insights include: **AI as Operating System:** Treat AI the way organizations treated the Internet in the 1990s — not as a technology upgrade but as a full redesign of how value is created and delivered. This means revisiting annual strategy cycles, multi-year planning horizons, and annualized budgets, all of which are too slow for AI-era competition.; **Proof-of-Concept Scaling:** Most AI pilots fail to scale because they solve narrow functional problems rather than representative organizational ones. Select problems large enough to signal broader transformation potential but small enough to deliver value quickly — the automotive example of compressing an 18-month car redesign cycle down to 18 weeks illustrates this sweet spot.; **Anti-Linear Strategy Design:** The single most common strategy killer is sequential, siloed thinking — corporate strategy feeding finance feeding marketing feeding operations. AI value emerges from connecting data across historically separate functions, such as using sales and marketing datasets to solve manufacturing problems, requiring intentional cross-functional data architecture from the start.

What did Nigel Vaz and Amy Bernstein discuss on HBR IdeaCast?

Publicis Sapient CEO Nigel Vaz, speaking at HBR's Strategy Summit 2026, argues that AI functions as a business operating system rather than a technology tool, requiring organizations to redesign decision-making speed, business models, and cross-functional data flows before competitors do. Key topics include: **AI as Operating System:** Treat AI the way organizations treated the Internet in the 1990s — not as a technology upgrade but as a full redesign of how value is created and delivered. This means revisiting annual strategy cycles, multi-year planning horizons, and annualized budgets, all of which are too slow for AI-era competition.; **Proof-of-Concept Scaling:** Most AI pilots fail to scale because they solve narrow functional problems rather than representative organizational ones. Select problems large enough to signal broader transformation potential but small enough to deliver value quickly — the automotive example of compressing an 18-month car redesign cycle down to 18 weeks illustrates this sweet spot..

How long is this episode of HBR IdeaCast?

This episode is 30 minutes long. SignalCast provides an AI-generated summary so you can get the key insights in about 3 minutes.

HBR IdeaCast

Strategy Summit 2026: Why AI Transformation Needs a Human Touch

March 12, 2026

30 min episode · 2 min read

Nigel Vaz,Amy Bernstein

Episode

30 min

Read time

2 min

Topics

Startups, Leadership, Design & UX

AI-Generated Summary

Published Mar 15, 2026

Key Takeaways

✓AI as Operating System: Treat AI the way organizations treated the Internet in the 1990s — not as a technology upgrade but as a full redesign of how value is created and delivered. This means revisiting annual strategy cycles, multi-year planning horizons, and annualized budgets, all of which are too slow for AI-era competition.
✓Proof-of-Concept Scaling: Most AI pilots fail to scale because they solve narrow functional problems rather than representative organizational ones. Select problems large enough to signal broader transformation potential but small enough to deliver value quickly — the automotive example of compressing an 18-month car redesign cycle down to 18 weeks illustrates this sweet spot.
✓Anti-Linear Strategy Design: The single most common strategy killer is sequential, siloed thinking — corporate strategy feeding finance feeding marketing feeding operations. AI value emerges from connecting data across historically separate functions, such as using sales and marketing datasets to solve manufacturing problems, requiring intentional cross-functional data architecture from the start.
✓Unit-Economics Measurement: Replace milestone-based project reviews with continuous micro-metrics. Track cost per software release, cycle time per feature, and defect escape rate rather than waiting for annual budget cycles to validate strategic hypotheses. These granular signals allow real-time inference about whether transformation is moving in the right direction.
✓Ethics Embedded in Technology: Responsible AI requires translating ethical principles directly into technical decisions — specifying which data leaves the organization, choosing between open-source and closed models based on training transparency, and enforcing data sovereignty rules. Employees using public AI chatbots with customer data because internal tools are inadequate represents a preventable, still-common failure mode.

What It Covers

Publicis Sapient CEO Nigel Vaz, speaking at HBR's Strategy Summit 2026, argues that AI functions as a business operating system rather than a technology tool, requiring organizations to redesign decision-making speed, business models, and cross-functional data flows before competitors do.

Key Questions Answered

•AI as Operating System: Treat AI the way organizations treated the Internet in the 1990s — not as a technology upgrade but as a full redesign of how value is created and delivered. This means revisiting annual strategy cycles, multi-year planning horizons, and annualized budgets, all of which are too slow for AI-era competition.
•Proof-of-Concept Scaling: Most AI pilots fail to scale because they solve narrow functional problems rather than representative organizational ones. Select problems large enough to signal broader transformation potential but small enough to deliver value quickly — the automotive example of compressing an 18-month car redesign cycle down to 18 weeks illustrates this sweet spot.
•Anti-Linear Strategy Design: The single most common strategy killer is sequential, siloed thinking — corporate strategy feeding finance feeding marketing feeding operations. AI value emerges from connecting data across historically separate functions, such as using sales and marketing datasets to solve manufacturing problems, requiring intentional cross-functional data architecture from the start.
•Unit-Economics Measurement: Replace milestone-based project reviews with continuous micro-metrics. Track cost per software release, cycle time per feature, and defect escape rate rather than waiting for annual budget cycles to validate strategic hypotheses. These granular signals allow real-time inference about whether transformation is moving in the right direction.
•Ethics Embedded in Technology: Responsible AI requires translating ethical principles directly into technical decisions — specifying which data leaves the organization, choosing between open-source and closed models based on training transparency, and enforcing data sovereignty rules. Employees using public AI chatbots with customer data because internal tools are inadequate represents a preventable, still-common failure mode.

Notable Moment

Vaz describes how an automotive client used iterative AI-driven market feedback during the design phase to determine whether to include a reversing camera — discovering that including it made the vehicle unaffordable in Malaysia — allowing supply chain and pricing decisions to adjust before production began.

Know someone who'd find this useful?

Episode Transcript

AI agents are suddenly everywhere you turn. But without identity, you can't really control what they do. Okta helps you assign every agent a trusted identity, so you get the power of AI without the risk. Secure every agent. Secure any agent. Okta secures AI. A new digital reading experience from Harvard Business Review is here. It's the HBR interactive issue. Swipe through pages, search each issue, and listen to articles with audio narration. The interactive issue is available now to all HBR Print Magazine subscribers. Not yet a subscriber to the HBR Print Magazine? Subscribe today at h b r dot org slash interactive issue. I'm Adi Ignatius, and this is the HBR IdeaCast. A few weeks ago, Harvard Business Review hosted a day long event looking at the cutting edge of strategy research and practice, the HBR Strategy Summit twenty twenty six. The day was filled with expert advice and guidance from both executives and academics. And for the next four Thursdays, we'll be sharing some of the best conversations with you on Ideacast. First up, a conversation between HBR editor in chief Amy Bernstein and Nigel Vaz, the CEO of Publicis Sapient. The company is in the digital transformation business, helping organizations modernize and adopt artificial intelligence to their existing models. That means he has had a front row seat to digital transformation at all kinds of organizations, and he shared his thought on what companies really need to do now around AI before it's too late. You'll hear him argue why AI should be thought of as an operating system, not a tool, how linear thinking is holding leaders back, and the most exciting opportunities he sees AI offering now. Here's that conversation between Amy Bernstein and Nigel Faz. You have had a ringside seat for strategy making all over the place, all over the world, many different kinds of companies. You have been doing it for years, so you have the long view. How has AI affected all of that, all the strategy making, all the thinking about strategy? If you could sort of boil it down. Look. I think AI is far more an operating system for how a business needs to operate than it is a technology. Right? Because I think we're at the beginning of a fundamental transformation where AI has been talked about as a technological trend, but it fundamentally is reshaping how businesses create and deliver value, much like the Internet did in the nineties. So for me, it's not so much about how AI is changing the process of strategy, but it's more how AI is changing how decisions are made and how work gets done. And if you think about how decisions are made and how work gets done evolving, then very quickly, you are, you know, having to change very simply the tempo of strategy. Right? So, you know, do annual strategy cycles work? Do planning shifts, you know, come in long multiyear cycles? Do budgets, you …

Get the full transcript (5,149 words) + summary by email — free

One-time email with the complete transcript and AI summary of this episode. No account needed.

One email, no spam. We’ll also show you what SignalCast does.

Browse all HBR IdeaCast transcripts →