Eye On A.I. Podcast Episode #322

Guest: Amanda Luther (Senior Partner, BCG; Global Lead, AI Transformation Practice)
Host: Craig S. Smith
Title: The Widening AI Value Gap (Inside BCG's AI Research)
Date: February 19, 2026

Episode Overview

This episode dives deep into BCG’s latest longitudinal research on the impact of AI across industries. Senior Partner Amanda Luther unpacks the "widening value gap" between early AI adopters and laggard companies. The conversation covers how and where top enterprises capture value from AI, the rise and limits of agentic systems, investment trends, and the challenge of scaling adoption company-wide. Listeners get a look behind BCG’s data collection and analysis, plus candid advice for organizations on starting and succeeding in the AI transformation journey.

Key Discussion Points & Insights

1. The "Widening Value Gap" in Enterprise AI

[01:17 – 02:15]

• BCG’s annual AI impact study, surveying 1,000–1,500 companies globally, highlights that the gap between AI leaders and laggards is increasing.
• Companies that invested early in AI enjoy a virtuous circle: AI delivers value, which they reinvest to achieve more impactful initiatives.
• “The companies that…were investing to begin with got value from that. That value flowed through into the P&L. They reinvested part of that value into additional tech and AI investments. And now they're getting more value from doing that.” — Amanda Luther [01:39]

2. Where Value Emerges

[02:41 – 04:14]

• 70% of AI’s value comes from core business functions: sales, marketing, procurement, supply chain.
• Remaining 30% is from corporate functions: finance, HR, general efficiency.
• The specifics vary by industry; e.g., marketing and R&D in CPG (consumer packaged goods), store operations, marketing, and labor in retail.

3. Types of Companies Realizing Value

[04:14 – 05:32]

• BCG segments companies by AI maturity:
  • 60%: Laggards/emerging—little to no AI value.
  • 35%: Scaling companies—seeing results in some functions.
  • 5%: "Future Built" leaders—AI value realized at enterprise scale, reflected in margins, revenue, and shareholder return.
• These leaders include both new digital-natives and century-old firms reinventing themselves.

4. Scale, Size, and the AI Accumulation Effect

[06:19 – 07:28]

• Large enterprises can accrue incremental time/cost savings across thousands of employees and invest in AI platforms with fixed costs.
• AI-native startups, though small, achieve high efficiency—e.g., $100M revenue with ~30 staff.

5. AI Value Measurement & the Virtuous Cycle

[09:31 – 12:37]

• BCG’s approach combines:
  • Surveyed maturity across 41 AI capabilities,
  • Direct questions about realized AI business value,
  • External financial data (EBIT, revenue, TSR).
• Positive correlation between high AI maturity and superior business outcomes.
• The gap widens as leaders reinvest 2x more in AI than laggards—a classic compounding effect.

6. Investment Numbers & Cost Reductions

[12:57 – 14:16]

• "Future built" companies invest double in AI (relative to peers), though average spend is still just 5% of IT budgets and growing.
• Leaders see up to 40% greater cost reductions than laggards, stemming from time savings, reduced vendor costs, and better procurement.

“The average global share of IT budget spent on AI is still only around 5%. It’s not a massive number… but it is a growing and meaningful number.” — Amanda Luther [12:57]

7. The Role and Reality of Agentic Systems

[16:05 – 20:08]

• Agentic systems currently drive about 17% of value for leaders—expected to grow to 30% in three years.
• Definitions of "agent" vary, so reported figures should be taken with caution.
• Effective applications are rarely “set-and-forget;” often require human-in-the-loop, thoughtful workflow design, and comprehensive context/guardrails.

“One very rarely is this a set it and forget it agent just runs. More often it is still kind of a human in the loop in the process making the final decision. But having an agent or a set of agents speed up some of the processes and take some of the toil out of the day to day.” — Amanda Luther [17:06]

8. Notable Agent Use Cases

[20:08 – 22:27]

• Marketing orchestration, customer contact centers (support and sales), and weekly business reviews in retail.
• Agents help “prep” and summarize complex data, but humans still drive strategic decisions.
• Applications are expanding into corporate functions like finance and HR.

9. Building vs. Buying Agents

[22:55 – 24:06]

• Only 11% of organizations primarily build agents themselves; most pursue a hybrid strategy using off-the-shelf solutions, major tech partners, and some custom builds for core advantage.
• Most advanced firms build agentic platforms to enable broader, low-code/no-code development internally.

10. Investment Breakdown: Tech, People, Process

[24:21 – 25:51]

• AI investment is broadly split: 50% on technology/platform/data, 50% on people.
• Leaders invest 6x more in upskilling/training than laggards (notably, in-person and real workflow-integrated).
• Process redesign and change management are critical: companies that combine tech with people/process get more value.

11. The Longevity and Strategic Value of AI Investment

[25:51 – 29:08]

• The value gap is likely to persist—or grow—as leaders reinvest AI gains.
• However, it’s not too late for "fast followers": monitoring, adopting proven solutions, and building organizational change capacity remain viable strategies.
• Companies that ignore AI risk withering.

“What you can't do is stay still, ignore this, put your head in the sand... The value gap is widening. You've got more to reinvest…” — Amanda Luther [27:22]

12. Adoption Patterns: Why Are So Many Companies Slow?

[30:27 – 33:54]

• Only about 5% of firms are truly "AI-first" at scale.
• However, the cohort just below—"scaling" companies—has grown from 22% (2024) to 35% (2025).
• 85% of companies now say they are doing something in AI, but only a few have made the full mindset shift to "what would an AI-first company do?"
• Changing entrenched processes and practices remains a major challenge.

13. Organizational Focus and the Challenge of Legacy Systems

[34:45 – 36:58]

• Even advanced companies like BCG prioritize AI transformation where it matters most to the business (e.g., client delivery), while long-tail internal processes lag.
• Legacy systems and sunk costs make overhauls gradual and opportunistic—often triggered by contract renewals or system replacement cycles.
• Focusing limited resources on fewer, higher-value activities yields better returns.

14. BCG’s Research Methods & Internal AI Use

[38:23 – 40:00]

• Surveys are typically filled out by C-suite or direct reports, sometimes supplemented across business functions.
• BCG increasingly adds AI layers for analysis, summarization, and data querying, both for research and client interaction.

15. Advice for Laggards: How to Start

[40:07 – 43:47]

• Begin with C-suite-driven vision and leadership.
• Empathize with the real adoption barriers (data, systems, talent shortage).
• Identify one or two truly strategic, high-value use cases—don’t try to do too much at once.
• Build a cross-functional team, trumpet early successes, and build momentum.
• Don’t get "paralysis by analysis" choosing the perfect solution; most vendors are "pretty good," and execution matters more than theoretical best-in-class.

“It is better to have a B solution that you implement in an A plus way than to get the A solution that you never actually implement.” — Amanda Luther [43:11]

16. SME Adoption & Consulting Landscape

[43:47 – 46:00]

• BCG focuses on companies large enough to justify its fees (aiming for 10x ROI), but is exploring more tool/AI-based offerings that could, in the future, serve smaller clients.
• The business consulting industry as a whole is ripe for some degree of AI-driven productization and disruption.

17. Talent Implications & Future Workforce

[46:55 – 48:38]

• Expertise in AI-powered business tools will be increasingly valuable, much like Salesforce skills in prior decades.
• Younger employees (“native” with these tools) may drive positive disruption and upskilling in established organizations.
• BCG sees its youngest hires already pushing use of new tools and workflows internally.

18. Continuing & Evolving the Research

[48:45 – 50:10]

• BCG updates the longitudinal AI study every year, keeping ~80% of the sample consistent.
• Data is increasingly used not just for benchmark reports but for customized client insight—an area where interactive AI analytics is growing.

Notable Quotes & Memorable Moments

“10 minus 9 still equals 10. Basically what they found was they took the process down to one day, but...the review cycle still was nine more days.”
— Amanda Luther on real-world limits of partial AI process automation [18:49]

“If you look at the overall breakdown of the numbers, it's probably half and half true tech and data cleansing platforms, your licensing costs, all of that and then half. On the people side... The AI leaders are investing six times as much in training and upskilling as the AI laggards.”
— Amanda Luther [24:21]

“There is an organizational muscle around being able to change and being able to adopt new technologies and adopt new processes. Like building that muscle, I think is incredibly important if you want to continue to survive and thrive in a world that changes faster and faster every day.”
— Amanda Luther [28:06]

“It is better to have a B solution that you implement in an A plus way than…the A solution that you never actually implement...this is all about adoption, process change, the human side.”
— Amanda Luther [43:11]

“By and large this is actually able to create more joy in someone's day and in someone's job. There's places where AI can reduce the toil...and actually allow you to spend time on the creative parts.”
— Amanda Luther [51:24]

Timestamps for Key Segments

• [01:17] Intro to BCG’s AI study and value gap concept
• [04:14] Maturity segmentation: laggards, scaling, leaders
• [09:31] How value is measured; correlation vs. causation
• [12:57] Investment levels and IT budget share
• [16:05] Impact and definition of agentic (agent-based) systems
• [20:08] Agent applications in marketing, contact centers, and weekly operations
• [22:55] Build vs buy: how companies source agentic solutions
• [24:21] Investment breakdown—tech vs people/training
• [25:51] The durability of AI investment: long-term widening gap
• [30:27] Current adoption rates and the surprising slowness in customer service
• [34:45] Why some basic business functions lag in automation
• [38:23] How BCG collects and compiles survey data
• [40:07] Practical advice for laggards starting AI adoption
• [43:11] The human/process side matters more than tool perfection
• [46:55] Talent trends; upskilling for the future
• [48:45] Annual AI study cadence and evolving analytics with AI
• [51:05] The human dimension: AI makes work more rewarding when properly applied
• [52:47] The real investment leaders make in in-person, workflow-integrated training

Conclusion: The Big Picture

BCG’s survey data shows that early and determined adopters of AI are reaping outsized business benefits, pulling further ahead of their peers. But Amanda Luther’s message is pragmatic and encouraging: it’s still early—companies don’t need to catch every wave, but standing still is not an option. The recipe for success is focused, C-suite-driven priorities, investment in both people and technology, and a willingness to experiment and learn by doing. Most of all, organizations should recognize AI as a long-term strategic lever—one with both business and cultural ramifications for those willing to lean in.