The world of Artificial Intelligence is evolving at breakneck speed, shifting from basic automation to intelligent, independent agents capable of executing complex tasks. As a portfolio CPO working with private equity backed companies across the UK, Europe, and the USA, I see navigating this transformation as more than just about keeping up - it’s about actively shaping the future of your product, team, and company.

I see this shift unfolding in six distinct stages, each representing a fundamental change in how AI operates and integrates into our businesses. Think of it as a progression from AI as a ‘co-pilot’ to AI as a fully-fledged collaborator, akin to moving from playing an instrument to conducting a full orchestra. Those who fail to adapt risk being left behind, as AI-powered organisations redefine what it means to innovate and compete.

Stage 0: The Pre-Chatbot Era – Early Automation

Before the chatbot boom, automation was largely rule-based. Expert systems, decision trees, and early machine learning applications (going as far back as the 1950s!) offered structured but rigid workflows. These systems worked well for predictable tasks but lacked the adaptability and learning capabilities we expect from AI today. The real turning point came with the explosion of large language models and conversational AI, setting the stage for more dynamic, context-aware interactions.

Stage 1: The Chatbot Revolution – AI Becomes a Familiar Face

For many companies, their first real encounter with AI came through chatbots. Tools like OpenAI’s ChatGPT, Google Gemini (initially called Bard), and Perplexity transformed expectations overnight. But despite their impressive language abilities, they were still limited - more reactive than proactive. Businesses started experimenting with ‘prompt engineering’ to coax better results, but it quickly became clear that general-purpose AI wasn’t enough.

Stages 1-2

Most companies are still in this stage.

Stage 2: AI as an Expert – The Rise of Specialist Agents

As businesses pushed beyond generic AI, domain-specific AI specialists emerged. These are AIs trained on focused datasets and tuned for industry-specific applications.

For example:

  • Juro and Robin AI are revolutionising legal contract analysis

  • Callsign applies AI to fraud prevention and digital identity verification

  • UiPath enables businesses to create their own AI-powered automation agents

These AI agents are no longer just chatbots; they act as embedded workflow partners, delivering real business value.

Stages 2-3

Experimental companies are at this stage now.

Stage 3: AI Agents Take Action – Autonomous Execution

Until recently, AI served mainly as a decision-support tool, providing insights but requiring humans to take action. That changed in 2022, when AI agents started demonstrating real autonomy.

Early breakthroughs include:

  • Devin (from Cognition AI), an autonomous software engineer that can develop, test, and debug code

  • SWE-agent, an open-source project proving that LLMs can become independent coders

  • AI-powered customer support bots that can not only answer queries but resolve issues end-to-end

The hype around generative AI and autonomous agents has led to an influx of startups exploring this space. The shift from ‘co-pilot’ to ‘executor’ is happening now.

Stage 4: AI as the Innovator – Creative and Strategic Thinking

Once AI agents consistently execute tasks, the next frontier is innovation. Future AI systems won’t just follow orders; they’ll generate ideas, optimise strategies, and push the boundaries of what’s possible.

Imagine an AI-driven product development team:

  • You set a goal (“increase user engagement by 20%”)

  • AI agents analyse the data, test new approaches, and refine the best solutions

  • Your role shifts from managing execution to guiding strategic direction

Companies like DeepMind (now owned by Google) and Nvidia are supporting AI-driven creativity and problem solving in fields like scientific discovery, gaming, and business intelligence. As these capabilities mature, AI will play a pivotal role in shaping product roadmaps and business strategies.

Stages 4-5

Pioneering companies are at this stage.

Stage 5: AI-First Companies – When AI Runs the Show

At this stage, AI moves beyond being an enabler to becoming the core engine of an organisation. AI-first companies will be structured around autonomous AI teams that make decisions, optimise processes, and drive execution.

For instance:

  • Ocado has automated warehouse logistics with AI-powered robotics

  • AI-driven trading firms are already making split-second financial decisions with minimal human intervention

  • Autonomous customer service teams are resolving complex issues with AI-driven troubleshooting

This isn’t some far-off sci-fi concept. It’s already beginning to take shape.

Stage 6: The Symbiotic AI Organisation – Humans and AI as Partners

The final stage envisions a world where humans and AI don’t just coexist but co-create. The distinction between ‘human-driven’ and ‘AI-driven’ disappears as organisations fully integrate AI into decision-making, execution, and innovation.

In this future:

  • AI handles routine and data-heavy decision-making

  • Humans focus on creative vision, ethics, and high-level strategy

  • Work-life balance could improve as AI offloads tedious tasks

While this vision is still emerging, the key question is: can businesses adapt fast enough to unlock the benefits without being disrupted in the process?


What Should Leadership Do Now?

If you're a Chief Product Officer, the AI agent revolution is already knocking on your door. Here’s how to stay ahead:

  1. Educate yourself and your teams. Beyond the AI headlines, understand the real implications for product development, customer experience, and business strategy

  2. Identify where AI can drive value and ask are there manual processes that could be automated / could AI agents help personalise customer experiences or accelerate product iterations?

  3. Experiment and iterate by starting small and running AI-powered pilot programs to learn from early adopters (success in AI is as much about culture as it is about technology)

  4. Develop complementary skills to prepare for a world where AI will handle execution, but human strengths like critical thinking, strategic vision, and emotional intelligence will become even more valuable

  5. Stay adaptable, accept that AI is evolving rapidly, and ensure your product strategy today may look completely different in two years (flexibility and a growth mindset will be key)

Final Thought: Are You Ready?

The rise of AI agents might seem daunting, but it also presents an unparalleled opportunity. Businesses that embrace AI will unlock new efficiencies, capabilities, and market advantages.

The question isn’t whether AI agents will dominate - it’s how prepared you are to thrive in this new era. The future of AI-driven product management is happening now. Will you lead the charge, or will you be left trying to catch up?

Previous
Previous

Why the Smart Bet Might Be on Small Language Models - Not LLMs

Next
Next

The AI 'Tsunami': It’s About Riding the Wave, Not Drowning In It