By: Max Ricciardo

As companies adopt transformational AI technologies, they’re simultaneously confronting a persistent challenge of change: trust. How can employees, customers, and even the companies themselves trust machine intelligence if they don’t understand it?

Transparency is a long-standing tool of success, yet many companies apply it inconsistently or without clear intent. Some leaders fail to shape coherent goals, close off the decision-making process, or discourage sincere feedback. Introducing AI into this mix has amplified uncertainty, particularly when employees don’t understand how leaders make decisions or how these technologies affect their futures.

That’s why AI experts recommend that companies build transparency into their operations and processes. Steven Kawasumi, a California-based AI product and technology executive, encouraged leaders to align AI transparency and leadership transparency toward similar goals: to create trust and alignment in complex systems.

“Transparency matters in everything involving AI,” Kawasumi said. “It helps people understand what inputs matter, how decisions are formed, and where a system’s limits are. The same goes for transparency in leadership. When employees understand the ‘why’ behind decisions, they execute better.”

Kawasumi has led large-scale AI and machine learning initiatives at Fortune 100 companies, focusing on how transparent decision-making builds trust in complex organizations. As companies evolve, the best leaders manage dynamic times with transparency. Kawasumi highlighted five tested techniques that remain applicable to the AI-enabled workplace.

Define Project Goals Beyond SMART

Companies devote significant time and bandwidth to goal-setting, which they ultimately waste without firm intent behind it. Goals provide reference points for decisions and tradeoffs, and the goal-setting process benefits from structure.

Many companies rely on the SMART method to create clear, measurable goals. While effective, the framework often falls short in complex, multi-stakeholder environments. Processes grow complicated when companies manage competing interests, multiple KPIs, and projects without a fully formed destination. Friction emerges when departments optimize for competing objectives, a problem compounded when intelligent systems are asked to operate against unclear or conflicting goals.

But for large organizations with multiple interests and stakeholders, SMART goals can be too simple. Kawasumi suggests that these companies take the next step to establish firm, consistent goals and clear metrics for achieving them.

“Larger organizations struggle when they manage competing interests instead of a shared objective,” Kawasumi said. “Their biggest issues occur when goals aren’t coherent. Clarity and consistency are important to make sure everyone is rowing in the same direction.”

Open the Decision-Making Process

Too often, leaders make high-impact decisions with limited input, even as organizations grow more complex. Business growth coach Bill Flynn has noted that organizations eventually outgrow individual leadership heroics, requiring systems that enable shared strategic decision-making.

Businesses now require systems that support shared decisions, which these technologies enhance. Yet Kawasumi noted that leaders waste that power when they silo decisions unto themselves. They also risk misunderstanding personal biases, alienating (and losing) talent, and even burning out from decision-making stress.

“Shared decision-making aligns teams more effectively, particularly when they disagree about a potential outcome,” Kawasumi said. “Leaders certainly should use AI to help make decisions. But they also risk undermining those decisions by failing to engage the talent they employ for the same purpose.”

Set Aside Time and Place for Feedback

Effective feedback systems help employees feel seen, heard, and aligned with organizational goals. According to the research firm Gallup, 80 percent of employees who report receiving meaningful feedback also say they’re more engaged in the workplace. That includes employees who work remotely.

Kawasumi pointed to several examples of feedback strategies that have proven successful. For instance, Glassdoor offers private Company Bowls for employees to provide input, anonymously or by name. The Society for Human Resources Management recommends that leaders initiate feedback with specific, objective thoughts; encourage questions, and follow up with employees. 

In a recent example, author and professor Adam Grant advised CEOs that feedback is most effective when it signals trust in employees’ ability to meet high expectations. Applied thoughtfully, this approach can help organizations move from one-way feedback to a more constructive, ongoing dialogue.

“Feedback is most effective when it’s consistent, structured, and intentional,” Kawasumi said. “Leaders improve transparency by purposefully creating space for it.”

Be Vulnerable 

AI is reshaping the workforce, automating jobs from entry-level programmers to management. In a Reuters poll, 71 percent of respondents said they fear permanent job loss, and even Google CEO Sundar Pichai has suggested his own role could be among the easier ones to replicate.

In response to rapid, technology-driven change, some leaders default to confidence and control, worrying that acknowledging uncertainty will undermine authority. Kawasumi argues that this instinct often backfires in complex environments where ambiguity is unavoidable.

In organizations shaped by advanced technology, vulnerability means intellectual honesty. Leaders who acknowledge uncertainty build credibility and demonstrate the ability to learn and adapt, qualities employees expect when navigating complex systems.

“Leaders who acknowledge uncertainty dramatically advance their credibility,” Kawasumi said. “If you can be open about challenges, you help employees understand their roles more fully. It helps them stay grounded and focused on solving the right problem.”

Be Mindful of the ‘Transparency Paradox’

In 2012, a technological lifetime ago, Harvard researcher Ethan S. Bernstein introduced the concept of the “Transparency Paradox,” arguing that excessive visibility can undermine trust. Bernstein found that employer strategies designed to open the workplace actually undermine employee trust and inhibit performance.

In the post-COVID workplace, this “paradox” has faced further tests from messaging apps, keystroke loggers, GPS, and other technologies employers use to monitor employees under the guise of “transparency.” Kawasumi said that applying monitoring tools for what amounts to tracking and bugging employees no longer represents a paradox.

“The idea of a paradox doesn’t go far enough. Today, I’d call it a fallacy,” Kawasumi said. “When employers misuse these technologies, they’re weaponizing the concept of transparency. And when you use the idea of transparency to be a bad actor, you have a problem.”

Ultimately, AI transparency and organizational transparency run parallel courses. AI architects can make their product more trustworthy by opening model-training procedures for inspection. Similarly, leaders can increase trust by bringing employees into the decision-making process.

Recently, the phrase “Clear is kind” has become widely cited regarding transparent leadership. However, Kawasumi emphasized that transparency works only when it helps people understand how and why leaders make decisions. Clear goals matter, but transparency only works when those goals and the reasoning behind them are shared across the organization.

“Transparency intrinsically matters to every organization,” Kawasumi said, “because it guides people to understand how decisions are made and the forces that drive them.”