Adobe's AI expert Rochelle Tognetti advises companies to move beyond small AI experiments toward integrated, governed systems. With over 25% of consumers already using AI agents for research, businesses must now prioritize data security and systematic governance to scale AI use safely. This shift is critical for companies looking to protect against operational risks while moving AI from simple task automation to independent workflows.
Enterprises currently stuck in the phase of isolated AI experiments must pivot toward building trusted, governed AI systems to realize genuine business value. Rochelle Tognetti, Adobe’s AI evangelist for the Asia-Pacific and Japan region, recently highlighted that while AI has become a household productivity tool, a major gap remains in how large organizations deploy it at scale. While many businesses are running successful task-specific pilots, these isolated wins often fail to translate into broader organizational efficiency.
Scaling AI Beyond Simple Automation
Tognetti noted that consumer habits have shifted rapidly, with more than 25% of global users now relying on AI chat agents to research products and make decisions. This creates a direct link between AI technology and brand reputation. Because AI is now positioned directly between businesses and their customers, companies cannot afford to treat it as mere software. Instead, they must reimagine entire business processes with the customer at the center, rather than simply automating existing, outdated workflows.
Governance and Risk Mitigation
The move toward more autonomous AI agents brings specific operational challenges. Tognetti warned that if these systems are not properly managed, they could create substantial liabilities, such as the accidental deletion of critical business data or incorrect customer communication. To mitigate these risks, organizations need to build foundations based on three pillars: trusted business context, strong internal governance, and strict security controls. These elements ensure that AI behaves predictably in live production environments.
The Four-Stage Adoption Roadmap
Adobe has proposed a four-stage framework to help companies structure their AI adoption. This journey begins with AI assisting employees and moves through augmenting human decision-making, automating workflows with human oversight, and finally, reaching a stage where AI operates independently. Currently, most enterprises remain in the first two stages. Adobe's approach involves placing a governing harness around large language models to ensure auditability and security. By integrating these systems into existing technology infrastructure, businesses can reduce the risks of unmanaged AI while working toward more complex automation. For investors and stakeholders, the key monitorable will be how quickly large enterprises can move from individual experimentation to these more secure, scaled frameworks, as this transition will likely determine the long-term return on capital invested in AI technology.