When Google unveiled what it called the "biggest upgrade" to Search in more than 25 years at its May 19 I/O conference, the company framed the rollout as the future of online search. The new product would offer AI-generated responses, conversational queries, and a redesigned Gemini-powered interface, replacing the familiar list-of-links experience users had relied on for decades. 

The reaction was immediate — and cold. Across Reddit, X and support forums, longtime users complained that one of the internet's most familiar products suddenly felt harder to control, less transparent and fundamentally different from the tool they had used for years. Competitor DuckDuckGo later reported increased traffic to its AI-free search option following the rollout.

For CIOs, the backlash reflects a challenge enterprises face constantly: how to modernize systems people depend on without disrupting the habits, workflows and trust that made those systems valuable in the first place.

Related:The power of CIO networking in the competitive AI world

Organizations go through versions of this every day. Companies redesign internal workflows, consolidate platforms, introduce AI copilots or replace legacy systems that employees have built their routines around for years. Leadership may view those moves as necessary modernization, but employees and customers often experience them as foundational disruption. Why the disconnect?

"The reason is actually relatively simple," said Simon Ratcliffe, fractional CIO at Freeman Clarke. "People rarely judge systems purely on technical merit. Rather they judge them on familiarity, reliability and how well they support daily routines."

Familiarity has business value

One of the recurring themes across enterprise transformation efforts is that organizations often underestimate how much value users place on familiarity itself.

Inside enterprises, employees build habits and expertise around systems over years, sometimes decades. Procurement teams know exactly how approvals move through internal systems. Warehouse workers develop muscle memory around logistics platforms. Finance teams create shortcuts and informal workflows around ERP systems. Over time, familiarity becomes inseparable from productivity.

"When something works, people stop seeing the tool — they see the outcome," said Mohit Ahuja, a strategy and transformation leader and consultant at Caterpillar. "The moment you change that interface or workflow, you've interrupted muscle memory they've built over years. That interruption feels like personal loss, not progress."

Related:Ignore the hype: Smarter tech bets at speed of change

That dynamic helps explain why even technically successful upgrades can generate outsized frustration. Organizations often approach modernization through the lens of efficiency or expanded capability, while users experience the same changes through the lens of routine disruption and lost competence.

Todd Nilson, a community and digital workplace strategist at Clocktower Advisors, said employees frequently interpret sudden workflow changes as a devaluation of the expertise they spent years building.

"Familiarity is a form of competence," Nilson said. "People build real expertise around the tools they use daily, and that expertise is tied to their professional identity. A forced update doesn't just change their workflow; it retroactively devalues the skill they've built."

Niel Nickolaisen, chairman of the CIO council at FC Centripetal and technology leader advisor at VCLM, said organizations often misjudge  how emotional change can be. He has observed that as a group, people instinctively reject the new thing simply because it is unknown.

"Humans prefer the familiar to the comfortable and the comfortable to the better," Nickolaisen said. 

The challenge for CIOs is that resistance to change does not necessarily mean employees oppose modernization itself. In many cases, they are reacting to uncertainty, loss of confidence, or the fear that leadership is changing systems faster than the organization can realistically absorb.

Related:How IT leaders build a culture for what comes next

The pressure to modernize keeps accelerating

At the same time, enterprise leaders face legitimate pressure to move quickly. Aging infrastructure, technical debt, cybersecurity concerns, competitive pressure and AI adoption initiatives all create strong incentives for modernization. The problem is that organizations increasingly attempt to execute several transformations simultaneously.

Ratcliffe said many companies now bundle platform migrations, workflow redesigns, AI rollouts and restructuring initiatives into the same programs — in the name of efficiency. But this approach misses the mark.

"While this may appear efficient on paper, it creates overwhelming levels of uncertainty for users," he said.

AI initiatives have intensified that pressure further because organizations fear appearing stagnant or technologically behind competitors. Several experts noted that AI-related change feels different from previous technology rollouts because it also affects how employees perceive their own professional relevance.

"Previous technology shifts changed where work happened or how it was processed," Ahuja said. "AI is changing who appears to be doing the thinking." 

That distinction matters because workflow disruption is easier to manage than identity disruption. Employees adapting to a new ERP system may feel frustrated; employees asked to work alongside systems that appear to replicate judgment, expertise, or creative work may experience something much closer to anxiety.

Nickolaisen said the compressed pace of AI adoption is amplifying those concerns and leaving organizations with less time to build trust and familiarity around the changes. Without strong communication and reassurance, he warned, organizations risk deepening employee resistance rather than accelerating transformation.

Change management failures are usually human failures

While each expert came to this topic with a different professional background, one point surfaced repeatedly: organizations continue to underestimate the human side of transformation. 

"The biggest mistake organizations make is treating change management as an afterthought," Ahuja said. "Technology teams spend 18 months building a solution, then allocate three weeks for training. That math never works."

The issue extends beyond training itself. Many organizations still treat rollout resistance primarily as an obstacle to overcome rather than a source of operational intelligence. 

"When employees raise concerns during rollout, they're frequently labeled 'resistant to change,'" Ahuja said. "That's a dangerous misread. Frontline workers often see failure points that no architect or consultant anticipated."

Nilson similarly argued that organizations often misread what adoption metrics actually indicate. Employees may technically use a new platform because they are required to, while simultaneously building workarounds, bypassing official workflows, or disengaging from the system entirely.

"Adoption without value-add is just compliance, and compliance is fragile," Nilson said. 

The more useful signal is the nature of employee resistance itself. Constructive criticism and detailed operational complaints often indicate users remain engaged enough to improve the system. Silence can be more dangerous.

"When frontline users stop raising issues, they've often stopped believing anyone is listening," Ahuja said. "That's when workarounds multiply and adoption numbers become fiction."

What CIOs should do differently

There is a strong argument that organizations need to treat modernization less like a deployment event and more like a long-term confidence-building process.

Ahuja described modernization as "a negotiation" between leadership urgency and operational reality. One approach he advocates is running old and new workflows in parallel long enough for employees to build genuine trust in the replacement system before cutovers occur.

Nickolaisen also emphasized the importance of familiarity-building before large-scale transitions. Early previews, pilot environments, optional testing periods and phased rollouts all help reduce resistance because users have time to develop confidence gradually, instead of being forced abruptly into unfamiliar systems.

For some, this is still too late in the process; Nilson believes users should be involved earlier in transformation planning itself.

"Don't build finished products and then ask for feedback," Nilson said. "Discuss the problem you're trying to solve before you've committed to a solution."

Metrics matter too, but it's critical to pay attention to the right ones. Evaluating transformation efforts solely through rollout deadlines or deployment completion will paint an inaccurate picture.  Ahuja recommended that CIOs should instead monitor employee confidence, workflow friction, customer satisfaction, error rates and the extent to which users actually trust the new system after rollout.

The broader lesson behind the Google search backlash is ultimately less to do with AI itself and more about how organizations approach change internally. Successful systems accumulate habits, shortcuts, trust and institutional knowledge over time. Those become part of the product experience, even if leadership teams stop noticing them.

For CIOs overseeing modernization efforts, the challenge lies in introducing change without undermining the confidence and operational stability that made people trust and engage with the system in the first place.