AI as an Engine for Growth: Moving Beyond Cost Savings

Artificial Intelligence (AI) is often framed as a tool for efficiency in terms of shaving minutes off tasks, automating repetitive tasks, reducing headcount, and generally cutting operational costs. While these benefits are real and cost-effective, this narrow view risks underutilising AI’s transformative potential for growth. 

AI should be seen not only as a cost-saving mechanism but essentially as an engine for growth which can drive innovation, create new markets, and redefine competitive advantage. In McKinsey’s 2025 State of AI, it was identified that the core leadership shift in 2025 is to treat AI as a market-making capability.

This shift is critical for policymakers, investors, and business leaders who aim to harness AI for long-term value creation.

The cost-saving narrative positions AI as a defensive strategy which can yield short-term gains but rarely creates sustainable differentiation. Competitors can replicate cost efficiencies, eroding advantage.

Moreover, focusing solely on savings often leads to underinvestment in AI capabilities that could unlock new revenue streams. High-performing companies that want to have a market advantage must target growth and innovation alongside cost reduction.

With AI, the opportunity for growth is tremendous. MGI estimates generative AI could add $2.6–$4.4 trillion in annual value across use cases such as customer operations, marketing, software engineering and R&D, which can directly expand revenue capacity. This approach aligns with historical patterns of technological disruption like electricity, the internet, and cloud computing, as they all drove exponential growth by enabling entirely new possibilities rather than merely reducing costs.

In practical terms, AI can be used to drive growth by creating market demand, which shifts internet search to discovery, thus increasing average order value and conversion. Amazon’s recommendation systems are reported to drive 35% of sales, a signal of how personalisation creates demand rather than just optimising funnels.

Cloud-native personalisation platforms (e.g., Amazon Personalise with Bedrock) now let firms re-rank content for explicit growth objectives.

Read also: AI Doesn’t Have a Trust Problem; It Has a Translation Problem

Netflix uses AI-powered recommendation engines not just to improve user experience but to expand global reach. By analysing viewing patterns, Netflix identifies regional content preferences, fuelling investments in local productions. This strategy transformed Netflix from a U.S.-centric service into a global entertainment powerhouse. AI enables companies to enter new markets by lowering barriers to personalisation and localisation. 

In terms of product innovation, AI enables companies to launch net-new offerings faster.  During the COVID-19 pandemic, Moderna leveraged AI to accelerate vaccine development. Machine learning models predicted mRNA sequences with high efficacy, reducing R&D timelines from years to months. This wasn’t cost-saving; it was market-making and revolutionary, enabling Moderna to capture unprecedented growth.

Another example is how AlphaFold’s evolution (AF2→AF3) moved from single-protein structures to complex interactions, broadening drug design and bioengineering. Through AI, manufacturing and commercialisation turn flexibility into revenue. BMW uses industrial AI across plants for quality assurance, logistics, and predictive maintenance, contributing to a highly flexible production network that can switch drivetrains on shared lines; a key to meeting dynamic demand for EVs without sacrificing throughput.

Real-time growth can be occasioned by AI through personalisation that increases customer lifetime value and opens cross-selling opportunities. Sephora’s AI-powered virtual try-on tools and chatbots enhance customer engagement, driving higher conversion rates and loyalty. These innovations create growth loops, where better experiences lead to more data, which in turn improves personalisation.

It would be right to also consider how PepsiCo, amongst the replete examples of how AI has been used to drive growth, partnered with AWS/Salesforce to build PepGenX, turning insights into faster product launches and scaled sales execution. This is a growth thesis: few pilots, more platformed capability.

Deploying AI as a tool for growth would undeniably have policy, investment and implementation implications for governments, large-scale companies with administrative bottlenecks and complex organisational structures, and, in fact, lots of players in the business space. 

Governments should incentivise AI adoption for innovation, not just automation. Tax credits and grants should prioritise projects that create new capabilities or markets. Regulatory frameworks must balance risk with flexibility, enabling experimentation in sectors like healthcare, technology and finance.

The updated OECD AI guidance (and related G7 frameworks) embeds risk management for general-purpose models, aiming for interoperability and diffusion beyond early-adopter sectors. Regulators should generally encourage policies that fund shared datasets and increase discovery capacity across small and mid-size firms.

In terms of investment, venture capital and corporate investment strategies should shift from ROI based on cost reduction to growth metrics covering market share expansion, new revenue streams, and customer acquisition. Investors should evaluate AI initiatives on their potential to create non-linear growth, not just incremental savings.

For workplace AI, studies around Microsoft 365 Copilot show ROI scenarios that include net-revenue gains and faster time-to-market; a reflection of commercialisation, not just “hours saved.” Business owners are encouraged to publish a growth Impact, profit & loss to keep track of the impact of AI investments on growth. Great AI depends on great data, and so executives must consider serious investment in data quality.

Image source: Unsplash

For implementation, Executives are advised to embed AI into strategic planning, not just operational efficiency. This involves:

• creating cross-functional AI teams that include product development, marketing, and strategy.
• measuring success through growth KPIs rather than cost metrics; and
• building scalable AI infrastructure to support rapid experimentation.

Lastly, operationalising AI as a growth engine requires cultural change. Leaders must champion AI literacy across the organisation, fostering a mindset that views AI as a creative partner rather than a threat to jobs. 

Nations that embrace AI for growth will outpace those that focus on automation. AI can drive GDP expansion through New Industries, Productivity gains in high-value sectors, and domination of emerging markets.

AI drives growth when leaders fund new delivery models with responsible governance as a precondition, not a postscript. The question isn’t “How much cost can we save?” It’s “What markets can we now enter, what products can we now design, and how fast can we scale them?” The organisations that answer those questions with data foundations, growth telemetry, and policy guardrails will convert AI into a flywheel for durable, compounding growth.