Using AI to Supercharge Enterprise Operations

Vishal Gupta sees AI as a fundamental force reshaping how products operate, how enterprises run, and how customers experience value at scale. For the former Global Chief Technology Officer and Chief Information Officer of Lexmark, artificial intelligence is deeply embedded across product engineering, customer engagement, and internal operational excellence.

Gupta shared how Lexmark is using large volumes of sensor data, generative models, and predictive systems to drive business impact, both within the company and for customers. He also reflected on where AI sits in the current technology lifecycle, how Lexmark built internal AI momentum, and the importance of measuring, not just adopting, AI.

How Lexmark Thinks About AI

Lexmark is more than a printer company; it is a technology company with a deep focus on IoT, cloud, data analytics, and AI. Gupta described the organization’s role as encompassing both CIO responsibilities (creating great experiences for employees) and CTO duties (building software that powers customer interactions).

On the pace of AI adoption, Gupta noted that we are in the early innings of this technology’s evolution. He reflects on adoption velocity across history, pointing out that while it took automobiles or cell phones years or decades to reach broad use, generative tools like ChatGPT hit 100 million users in a matter of months. This rapid adoption signals that AI’s integration into everyday enterprise processes is accelerating.

“If you think of ChatGPT, a hundred million users took two months… we do live a little bit in unprecedented times in terms of how fast the adoption … can happen.” But he cautioned that assessment of real, sustained value takes careful work and strong data foundations, not just hype.

AI Built on Data and Culture

Gupta emphasized that Lexmark’s ability to deploy AI widely was not accidental but the result of multi-year investment in data infrastructure and organizational alignment. The company built a data lakehouse that ingests and refines data from hundreds of systems, enabling deduplication, tagging, and strong analytics foundations for models.

That foundation enabled Lexmark to apply AI where high-value tasks already existed: for example, accelerating responses to customer RFPs. Because Lexmark had archived hundreds of successful responses, they could, “…train generative AI using the ones we want… That way we could generate good answers to the new RFPs… We think that will create almost $5 million in incremental benefits a year…” This illustrates how domain-specific data + generative AI = competitive impact, especially in scenarios where speed and scale matter.

Predictive Maintenance to Product Longevity

One of the most compelling examples Gupta shared arises from AI-driven maintenance and lifecycle optimization of hardware. Lexmark printers embed hundreds of IoT sensors that stream billions of data points into real-time processing systems. With machine learning models in production, Lexmark can now analyze streaming data and understand root causes of hardware failures, enabling them to extend printer life from about four years to seven.

Instead of merely reacting to failures, the company uses predictive insights to guide design decisions, supplier choices, and service processes, creating a substantial win for customers, the environment, and the business. This model of data-driven product longevity is not only operationally efficient but also strategically differentiating in a commodity hardware market.

Real-World AI Use Cases

Gupta shared several tangible AI use cases Lexmark has deployed, demonstrating both process improvement and customer experience transformation:

• Predictive Maintenance & Self-Service: AI models monitor supply levels and automatically trigger toner shipments before depletion, ensuring uptime without human intervention, and contributing to retention rates that now approach 99 percent.
• Edge AI for Service Experience: Lexmark is working with a large oil-change company to use edge AI that detects whether a vehicle has been greeted by staff within the first few minutes of arrival. If not, staff can intervene, reducing customer drive-offs. Gupta noted that this small improvement “pays for the entire month of the technology itself.”
• Personalized Sales Engagement: By applying AI to prospect data, including social feeds and product affinities, Lexmark has seen 2–3x improvement in response rates on targeted outreach versus traditional cold targeting.
• Generative AI in Customer Care: Lexmark is building AI solutions that can significantly reduce customer support handling times by interpreting historical data and real-time inputs, an effort expected to yield tens of millions in value once fully rolled out.
• AI-assisted Product Experiences: Lexmark has embedded AI assistant and coding assistant features into products. These allow customers to query how to perform tasks or generate SQL from plain language, giving users self-service capabilities that reduce reliance on professional services.

These examples show how AI can be integrated deeply into products, operations, and customer interactions, not just as bolt-on automation.

Measuring AI Beyond Completion to Accuracy and Trust

One theme Gupta reiterated, and felt many organizations overlook, is that metrics matter. “People underestimate the accuracy aspect of AI… not about ‘how do I get a project done,’ and not about ‘how do I actually measure accuracy on an ongoing basis to know whether or not I can really trust it.’”

He argues that accuracy and trust should be measured continuously, not assumed. Without objective measurement, companies can either leave competitive advantage on the table or expose themselves to risk. This distinction between deploying AI and operationalizing it with measurable confidence is increasingly critical as enterprises scale their AI footprint.

Democratizing AI Advantage

Looking forward, Gupta believes the ultimate promise of AI is democratization by enabling anyone, anywhere to achieve more with less friction: “What AI is able to do… is to level the playing field for any of us to achieve our ideas in a better, faster way and achieve our full potential.”

However, he also acknowledged the dual nature of AI: while it can empower individuals and organizations, it can be misused, such as for more effective phishing attacks. This means leaders must embrace safeguards and trust frameworks alongside innovation.

Lessons Learned

Across his conversation, Vishal Gupta shared several clear strategic lessons for CIOs and technology leaders:

• Invest in data foundations first. AI is only as effective as the data it is built upon and strong data architecture enables rapid, trustworthy deployment.
• Measure AI outcomes, not just outputs. Tracking accuracy and trustworthiness ensures real operational value and risk control.
• Embed AI deeply, not superficially. Use cases that integrate into products or processes have greater strategic impact.
• Scale talent and culture. Lexmark’s journey involved skilling existing talent and embedding AI fluency across teams: a model for other enterprises.
• Balance innovation and governance. Democratizing AI while safeguarding against misuse requires strategic guardrails and continuous learning.

Vishal Gupta’s perspective reveals that AI’s role in enterprise operations goes far beyond automation or efficiency gains. At Lexmark, it is transforming products, enhancing customer engagement, optimizing internal processes, and unlocking new business models.

Rather than treating AI as an isolated project, Lexmark has embedded it into the core of its technology strategy, from predictive maintenance to customer care to personalized customer engagement. Gupta sees this journey not as a fad, but as part of a long-term shift where AI becomes as ubiquitous in business operations as the internet or cloud computing.