Lessons from the Room: 5 Hard Truths About Closing the AI Value Gap

On the surface, it appears that every enterprise is effortlessly scaling AI to unlock massive productivity gains. Behind the scenes, however, the conversations reveal a completely different reality.

Global spending on AI has climbed past $250 billion and is forecasted to reach $2.59 trillion in 2026. Yet, comprehensive data from BCG revealed that only 5% of companies are achieving meaningful value from AI at scale, while 60% are seeing no material returns at all. As a leader at OpenAI recently observed, model capabilities are no longer the issue. Rather a massive gap has formed between what the models can provide and the actual value enterprises are extracting.

We’ve heard it firsthand in our conversations with leaders. This spurred a recent, private Breakthrough roundtable executive briefing hosted by Breakthrough CEO, Zannah Ryabchuk to unpack this disconnect with senior executives and CEOs across major enterprises. Instead of theoretical future capabilities, the conversations looked at the actual operational friction that occurs when fast technology collides with complex corporate realities.

A common thread that ran through the room is that leaders are not struggling with software deployment. They are struggling with a deep disconnect between individual employee habits and corporate strategic readiness.

If your organisation is wondering why a heavy investment in AI hasn’t transformed your bottom line, here are the 5 hard truths from the room.

5 Hard Truths from the Room

The executives at our Breakthrough roundtable exposed five distinct areas where execution must shift to close the value gap.

1. Shift the Traditional Approach

When 80% of your energy is spent on software licensing, cloud infrastructure and security measures, this leaves daily process integrations and workforce adoption as an afterthought. And as AI is accelerating faster than implementation can allow, it is paramount that you first focus on who will be the people, teams and workforce driving this transformation, implementing it and scaling it first, i.e. their current reality, their ways of working and their mindsets. Here sustainable value is created from reversing the entire traditional project pipeline.

💡High-achieving organisations, prioritise their people first, redesign the process second, and implementing the technology last. Doing it in any other order turns your adoption strategy into an exhausting, expensive game of cat and mouse.

2. Speeding Up a Broken Process Creates Faster Chaos

Driven by a fear of falling behind, leadership teams often rush to layer AI tools directly over legacy workflows, hoping the machine will magically clean up operational inefficiencies. There’s a misconception that the tech has the capability to clean up underlying inefficiencies. At Breakthrough, we’ve seen that AI actually acts like an accelerator. When layered over complicated, multi-step bureaucratic processes, it simply creates automated high-speed confusion.

💡 Executives revealed that most immediate value comes from strictly compressing daily cycles like repetitive tasks and simple customer FAQs, freeing up their people to handle more complex, specialised work. So simply put, never automate a broken step. Aggressively review and simplify these processes and data layout before AI is ever overhauled.

3. AI is a Mirror, Not a Bridge

The current challenge here is that management still views AI as a tool to bridge structural gaps or force collaboration across disconnected departments, particularly after an M&A. This is a fundamental misdiagnosis. Technology cannot fix broken communication lines. AI doesn't create a new culture. It aggressively amplifies your existing one. If your organisation has deep functional silos, it will expose this. If your workplace suffers from a low-trust culture, the rollout will fraction it further. The roundtable revealed two distinct behavioural risks happening right now:

1. As AI tools take months to approve, eager individuals who likely already utilise AI personally simply bypass management to build unmonitored process for efficiency and quick wins, exposing the business to data leaks.

2. Conversely, others are utilising the tech as a crutch. They blindly accept basic outputs without critical analysis, double checking facts or cross referencing.

💡Closing the readiness gap requires transparency and psychological safety.

It’s important for employees to feel safe to communicate errors, be clear on where AI was implemented and that leadership clearly understand the fears and concerns of their teams. Otherwise disconnect will continuously be a factor.

To uncover how you can build this psychological safety through feedback in your organisation, watch this Insight Short from Breakthrough Facilitator, Kate Arneil. 

4. AI Must Be Owned By All, Not just the CIO

When IT or procurement teams handle AI rollouts entirely, it inadvertently stalls the organisation's initial growth strategy. These departments are fundamentally wired for defence and protection. Therefore they default to priorising risk mitigation and security. This defensive framing causes workforces to fracture three distinct groups:

1. Employees who check out, passively waiting for step-by-step corporate handbooks before altering their routines.
2. Workers who retreat into silent resistance, quietly hiding their traditional processes.
3. Eager risk takers who bypass management entirely to build unmonitored shortcuts, fueling the rise of unmapped shadow AI.

💡Centralised training sessions move far too slowly to guide these groups effectively. The shift from resistance to adoption requires ownership and empowerment. AI must be owned by everyone from the top down and the bottom up. But an effective approach that works is having departmental champions of this transformation who better understand the daily role of their specific teams.

This reality is backed by a 2026 SAP enterprise study showing that 60% of businesses admit their employees lack proper AI training, which is fueling a surge where 68% of staff now use unapproved shadow AI tools just to get their work done.

5. Transformation Fatigue is Alive and Well

A clear warning from the room was the threat of pilot fatigue. Launching dozens of small, isolated AI trials across different business units might look productive but it rarely delivers scalable capital returns. Instead, it creates immense transformation fatigue among middle management, especially when layered on top of legacy digital transformation efforts from recent years.

💡Stop chasing breadth and start seeking depth. By abandoning scattered trials and focusing instead on small, highly concentrated operational adjustments, what we call One-Degree Shifts, you bring your team along without overwhelming them. As AI capabilities evolve almost weekly, typical annual training models are quickly obsolete.

Leaders must model and reinforce a learner’s mindset by visibly protecting dedicated windows of time each week solely for hands-on employee testing and peer-to-peer teaching. This single ritual has proven to radically lower workplace anxiety and spark organic innovation.

Conclusion

Since the GenAI boom, your growth strategy cannot rely on rigid, multi-year timelines. We are on a continuous, high-speed wheel of change, and AI has permanently disrupted how we evaluate baseline business performance. As you look to bridge the AI value gap in your own organisation, step back from the technical features and look at the cultural architecture.

Ask yourself: How have we directly linked our AI spend to our core business goals, and what specific steps are we taking to build a high-trust culture that is agile and resilient?

It isn’t an IT challenge to be solved in a silo by your CIO. It must be led directly by the executive team. Tools are easily replicated and are increasingly becoming more complex. A high-trust, adaptable culture that can navigate technological disruption is your only sustainable competitive advantage.

Ready to close your organisation's value gap?

Contact us today and align your leadership team and wider culture with technological acceleration.