AI-first, not AI-frantic: leading with purpose in the age of endless innovation.

In an age where AI innovation explodes daily - as evidenced by the Google's recent I/O conference in Mountain View, California and the Google Cloud Summit Nordics 2025 in Stockholm - business leaders face a critical dilemma: how to move beyond the hype and strategically integrate AI without descending into 'AI-frantic' chaos. Possibilities emerging from Google, Meta, Microsoft, and specialist AI companies like Anthropic and OpenAI leave many wondering: where do we even begin?" 

We are, in essence, in an 'interregnum' - a chaotic yet transformative gap where old paradigms of innovation struggle to adapt to the dizzying pace of new AI possibilities. In politics this means the gap between two parliaments or two leaders - someone has been deposed, but the new regime has yet to begin. It is a period of discontinuity, but shaped by the complexity of endless innovation.

The real question is how businesses today start using this myriad of AI tools. The options are changing daily. How do business leaders step back from the online debate about ‘the future of AI’ and start defining where the real use cases can be created in their business? In short, how should modern companies embrace the AI revolution?

There is a bottom-up argument that argues innovation should come from the people at the front-line of the business. These people are focused on the business and the solutions needed, they contend, so innovation should not be dictated by the IT leaders, such as the CIO. But look around at CIOs today. They are tired of business teams testing out AI solutions with no approval from IT. People are running pilots and experiments without any idea how these can possibly be scaled up into a production system - never mind how this can be budgeted for.  

At the same time, we can’t wait for AI to be mandated by executive leaders. By the time a policy is created and permission is granted for everyone to use a specific system, it will be months (or years) out of date. AI does need to be embraced within companies at all levels, and every business will have their own ideas on how best to encourage innovation in a way that respects the available budget for IT infrastructure.

However, I would argue that we need to focus on a cultural change in addition to worrying about budgets. This journey begins with fostering a truly permissive culture. It's about empowering every team member to ask, "Can we solve this problem or increase our productivity using AI?" and then giving them the green light to try. Encourage contained experiments: those that focus on individual or small team-led explorations, using readily available tools like Google Gemini or Notebook LLM, where the ideas are 'contained’ - meaning they don't require immediate oversight from compliance or legal, but rather focus on personal or team productivity gains. This hands-on experience is crucial for building widespread AI literacy and enthusiasm within your business.

I see this in action at Transcom every day, with hundreds of use cases demonstrating just how quickly AI is being adopted and the real benefits it brings. Our sales teams are using AI to create better client proposals. Our legal team harnesses it to check up on laws and stay compliant. Our marketing team uses AI to generate value proposition statements for refined customer personas. We’re also focused on upskilling our teams for this AI-driven world. For example, instead of spending most of their time manually reviewing recordings, our Quality Analysts (QAs) are now training AI systems, digging deeper into data insights, and pinpointing coaching opportunities.

AI is such a versatile tool. It has a place in every corner of every knowledge-based company.

You can’t make a claim to be an AI-first business if you are not using AI in every area of your business. So you need to start with this permissive culture. It’s OK to try these tools. Create your own experiments. See how they can improve your own work, so long as the ideas are contained and don’t require oversight by compliance or the legal team. But then actively work on an ideas funnel.

The cultural shift and encouragement to experiment take care of your first problem. To be an AI-first business means everyone on your team needs to experience what AI can do. But you also need to create a pathway for larger projects that go beyond what an individual can achieve.

This creates a pathway for grander ideas to get some internal seed funding and guidance from the IT team. This allows you to think bigger. To stop considering AI as “just a tool” or “just a service”.

AI is a tool, but it is a transformative tool that has the potential to create entirely new ways of delivering services. You may have a team member who suggests an idea that could transform your business or even comes up with a spin-off service that complements the main business. CIOs want to be creating and managing innovation, not clearing up the mess where business teams failed to think about how their pilot project can scale up. The fact is that CIOs are often caught between enabling experimentation and protecting the core architecture, data integrity, and budgets. Left unmanaged, a flood of AI pilots from every corner of the business creates more noise than progress.

Leaders need to focus on encouraging an AI-first culture, but only alongside the creation of the ‘ideas and innovation funnel’. This funnel, ideally managed by a collaborative committee involving IT and business leaders, acts as a filter and accelerator for promising AI initiatives. This will allow the best ideas to be taken forward and tested further in a digital playground - a sandbox environment insulated from the production systems - where promising concepts can be robustly assessed without risk.

Creating a “learn fast and fail fast” culture will allow these experiments to be fairly judged. What was the success criteria before the experiment was tried? Did it meet the objectives? Could it be scaled to change how your business functions? Encourage everyone in the business to suggest ideas that can go forward for testing, so long as you define what will be continued and what will be retired quickly. If the test is not working out, then scrap it and move on quickly.

Innovative companies have long embraced this idea, championing the belief that you get the best concepts and use cases by trying as many things as possible. You quickly retire those that don’t work and focus on those that are performing well. 

It is not yet clear which AI technologies will fundamentally change all businesses. Platforms and packages are changing daily. Smart company leaders will be building a culture that embraces experimentation using AI, but also building a pathway to real use cases so the ‘CIO fatigue’ is managed by ensuring that projects are either invested in or retired quickly.