Large enterprises’ swift adoption of generative AI (GenAI) is driven by the quest for cost reduction and operational efficiency—but leaders risk falling behind competitors if these short-term gains stagnate without long-term innovation capabilities.
This tension is particularly evident in industries that rely on customer interactions and agile processes. A joint study by Stanford and MIT found that companies using AI in customer service saw a 13.8% increase in resolved chats per hour, highlighting how AI can boost productivity and accelerate adaptation when integrated effectively.
Yet, many organizations confine GenAI to incremental improvements, missing its potential as a transformative force.
In this episode of the ‘AI in Business’ podcast, Deborah Golden, Deloitte’s Chief Innovation Officer, shares her perspective on how enterprises can use GenAI to unlock creativity, balance near-term efficiency with future innovation, and foster a culture of rapid experimentation.
This article explores three key insights from their conversation:
- Leverage diverse thinking styles for AI innovation: Bringing in team members with unconventional approaches to unlock creative potential and lead to transformative AI solutions.
- Balance immediate AI efficiencies with long-term innovation: Prioritizing projects that deliver quick operational gains in deterministic workflows while developing future capabilities around predictive enterprise capabilities ensures both near-term ROI and sustained growth.
- Foster a culture of AI-driven experimentation: Encouraging teams to “harness chaos” through short-cycle testing, cross-functional evaluation, and iterative scaling of AI systems accelerates learning and helps organizations stay ahead of technological change.
Listen to the full episode below:
Guest: Deborah Golden, Chief Innovation Officer, Deloitte
Expertise: Enterprise Technology, Digital Transformation, AI Adoption
Brief Recognition: Deborah leads Deloitte’s technology strategy and innovation efforts, advising Fortune 500 companies on adopting GenAI to drive operational efficiency and long-term growth.
Leverage Diverse Thinking Styles for AI Innovation
Generating real business impact with generative AI (GenAI) requires more than technical skill — it demands diverse teams capable of uncovering novel applications and reducing model blind spots. According to Deborah Golden, enterprises often overvalue only three to five types of thinking, unintentionally limiting AI’s innovation potential.
Golden emphasizes that this narrow cognitive spectrum prevents teams from challenging assumptions — a critical gap when developing AI systems that reflect the limitations of their training data. She advocates combining technologists, designers, business leaders, and unconventional thinkers to push the boundaries of AI’s role in operations and customer engagement.
“If you look across most enterprises, they often value just three to five types of thinking. But when you intentionally bring in individuals who think, learn, and problem-solve differently. You begin to see solutions and innovations that siloed teams would have missed.”
— Deborah Golden, Chief Innovation Officer at Deloitte
Business leaders driving GenAI projects should prioritize assembling teams that:
- Pair creative problem solvers with technical AI specialists to uncover non-obvious solutions.
- Integrate professionals from different industries to surface alternative use cases and reduce internal bias.
- Encourage nonconformists to challenge models and business processes, helping to spot hidden risks early.
Golden underscores that the best AI innovation happens at the intersection of creativity and technical expertise. Enterprises that intentionally diversify their AI teams will move beyond efficiency gains, unlocking new revenue streams and securing long-term competitive advantage as GenAI reshapes industries.
Balance Immediate AI Efficiencies With Long-Term Innovation
Focusing solely on short-term AI efficiency gains risks, leaving companies vulnerable as competitors scale GenAI into core business models. Deborah believes that the actual value of generative AI (GenAI) emerges when enterprises connect operational wins to long-term transformation.
Golden stresses that companies limiting AI to quick cost reductions often plateau, while those aligning near-term gains with strategic innovation build lasting competitive advantage.
“If you’re only looking at efficiency gains, you’re missing the broader opportunity,” says Golden. “The real value of GenAI emerges when you connect those early operational wins to bigger, organization-wide transformations.”
Golden recommends that business leaders structure their GenAI efforts across two parallel tracks, for example:
- Operational quick wins: Automate customer service tasks, enhance content creation, or reduce back-office costs.
- Strategic innovation bets: Develop GenAI-powered product features, explore dynamic pricing models, or automate demand forecasting and resource allocation.
This dual-path approach allows teams to demonstrate immediate ROI while building capabilities to unlock new revenue streams — such as AI-driven personalization in customer journeys or automating data-informed decision-making at scale.
Golden emphasizes that executives who treat GenAI as both a cost-efficiency tool and an innovation engine position their organizations to scale AI beyond isolated functions. This approach future-proofs the enterprise—ensuring it can evolve from cost-cutting into revenue-driving differentiation as GenAI capabilities mature.
Foster a Culture of AI-Driven Experimentation
Rigid processes and hesitation around AI experimentation expose companies to the risk of falling behind faster-moving competitors. According to Deborah Golden, traditional corporate structures often default to incremental improvements, limiting an organization’s ability to unlock GenAI’s full potential.
Golden stresses that successful AI adoption requires more than pilot projects — it demands a company culture capable of iterating quickly and scaling what works. Leaders must empower teams to operate in controlled testing environments where failures are treated as valuable learning cycles.
Golden recommends that executives foster this experimentation mindset by:
- Short-cycle testing: Rapidly prototype AI applications with small teams, measuring outcomes within weeks.
- Cross-functional evaluation: Engage business leaders alongside data scientists to assess pilots for both feasibility and business impact.
- Iterative scaling: Expand promising AI solutions gradually, refining them based on user feedback.
Companies that embed this adaptive experimentation approach accelerate their innovation cycles, reduce time-to-value, and position themselves to lead as GenAI reshapes market dynamics.
