Buckle Up: Navigating the Mind-Bending Acceleration of AI
Greg Montgomery, Principal Consultant at ClearPoint, delivered a keynote aimed at helping business leaders navigate the rapid changes AI is bringing, not through abstract forecasts, but through practical examples and actionable strategies. His message was clear: AI is not something to fear or admire from a distance, it’s something to use, test, and integrate into your business operations now.
Montgomery structured his talk around three concurrent “waves” of AI impact: operational efficiency, enhanced user experience, and entirely new business models. These, he argued, are not happening in sequence, they’re colliding at the same time. He illustrated this using a live experiment where an AI agent browsed Woolworths’ website, independently selected products based on a brief prompt, made substitutions, and scheduled delivery. This wasn’t a theoretical use case. It happened with publicly available tools and without the company needing to enable it. “That website was never designed for that kind of user,” he noted. “But the AI did it anyway.”
One of his strongest themes was the emerging gap between consumer AI tools and internal business systems. Employees and customers alike are now using apps like ChatGPT, Claude, and Gemini in their personal lives, often more powerful and responsive than the legacy tools inside their organisations. This growing disconnect creates both competitive pressure and cultural strain, as traditional systems begin to feel outdated.
Rather than advocating for a total overhaul, Montgomery emphasised adaptability. He encouraged businesses to develop an internal culture of experimentation, where small AI pilots are not only allowed but encouraged. “Be engineered for change,” he said. In this view, success with AI is less about finding a single breakthrough use case and more about building a portfolio of fast, iterative wins.
He closed by pointing to New Zealand’s history of innovation and problem-solving, framing the current AI moment as another shift that will reward action over analysis. “Go explore. Play. Learn. Ask, what if?” he said. “The only way to discover the limits of the possible is to go beyond them.
Takeaways
Employees are already using AI tools that outperform your internal systems.
This creates a capability and expectation gap. When your team can automate a task at home but not at work, frustration grows. Businesses should assess where employees are using external tools and consider enabling or integrating them safely into workflows.
Autonomous AI agents are now acting as digital users, without being invited.
An AI agent can browse your site, make decisions, and complete actions that your UX was never designed for. This shifts the paradigm: you’re no longer just designing for humans. You’re designing for human–AI hybrid workflows. Understanding how agents behave on your platform is now part of digital strategy.
Traditional performance benchmarks can no longer keep up.
AI models are improving so rapidly that evaluation tools created a year ago may not reflect their true capabilities today. Businesses need to revise how they assess AI tools, and potentially invest in continuous benchmarking strategies.
Solve real problems first, don’t get distracted by novelty.
Montgomery cautioned against using AI for its own sake. He showed how AI has been used to automate M&A due diligence, extracting insights from documents and generating summaries in minutes, saving significant time and cost. Identify repetitive, high-effort tasks, and start there.
Set your organisation up for small, rapid experiments.
Large-scale AI transformation projects can stall due to complexity. Instead, encourage small test projects with clear goals, cross-functional support, and room to fail. This builds internal confidence and allows teams to learn by doing.
AI will shift job roles and team dynamics.
As AI takes over the routine first 80% of a task (data entry, basic analysis, triage), humans will be expected to handle edge cases, judgement calls, or final oversight. This has implications for job descriptions, training, and team structure.
Leaders need to be hands-on with AI tools.
It’s no longer enough to rely on technical teams. Executives should personally try tools like ChatGPT, Claude, and Copilot to better understand their strengths and weaknesses. “You don’t need to be an expert,” Montgomery said, “but you do need to be familiar.”
Regular use is the best preparation for change.
Waiting for certainty will only widen the gap between your business and competitors. Using AI regularly, at any scale, builds experience, uncovers opportunities, and lowers adoption risk over time.
There is no final destination. AI adoption is ongoing.
Don’t treat AI like a one-off initiative. Instead, build processes that allow continuous learning and updating. Teams should expect to adapt as tools improve. “Be engineered for change,” Montgomery said, not just for now, but for what comes next.
