The race to real time in the age of AI

The race to real time in the age of AI

Read the key themes from our thought leaders’ panel and roundtable on 25th September 2025 in London.

What does it take to compete in the age of AI?

At our recent event in London, moderator R “Ray” Wang of Constellation Research explored this question with panel speakers Professor Savvas Papagiannidis from Newcastle University and the National Edge AI Hub, and Dolo Miah from Linebreak.  

The discussion was then opened up to a wider round table with researchers, academics, tech partners and enterprise leaders from sectors including Manufacturing, F&B, Financial Services and Insurance.  

Moving beyond the hype the participants explored how macro forces, decentralisation and AI are reshaping business - and why real time is no longer optional, but essential.

Discover five key themes that emerged - each one a challenge to rethink how we build, lead and compete in the age of AI.

1. Don’t underestimate the exponential impact of AI

Forget incremental change: AI is driving exponential efficiency. Ray explained:

“You need to be thinking in terms of 1000x in the Age of AI.  You need to be 10x faster, 10x better, and 10x cheaper or you’re out of the game”.

Ray sees a new wave of exponential efficiency potential:

• Level 1: 10x cheaper or  10x better

• Level 2: 10x cheaper and  10x better

• Level 3: 10x cheaper, 10x better and  10x faster = 1000x

These gains are leading to the rise of tiny teams: tech companies that are generating millions in ARR with just a handful of employees. This is the power of AI exponentials.  

This isn’t just about productivity gains. It’s about market dominance. There’s a new baseline: companies that build AI-first business models will outperform the market 10 to 1.

2. The age of AI is defined by closed, centralised systems: keeping control of your data is key

The Internet Era was characterised by broad ecosystems and many players, a decentralised mindset, open and interoperable to gain the widest adoption possible.  

However, today’s AI landscape is dominated by a few powerful players with closed expensive models and highly controlled, centralised systems. It’s a winner-takes-all world.

Ray described the current state as “AI arbitrage where companies leverage existing AI tools to deliver value without building their own capabilities. It’s fast but you’re dependent on someone else’s infrastructure, model and roadmap.”
‍
Dolo Miah considered:

“Are we just sleepwalking into another lock-in with AI? If we’re not careful, we’ll trade one centralised model for another. We need to keep control of our data and our decisions to shape our future.”

3. If your data is everywhere, your AI needs to be everywhere too

We’re in the zettabyte era. IDC predicts we’ll generate over 200ZB globally in 2025 with that number soaring to over 500ZB by 2029.  

Data is exploding everywhere, the vast majority of which is outside the cloud. It’s in the operational field e.g. across oil platforms, retail stores, supply chains, sensors and manufacturing robots etc. Dolo summed it up:

“If your data is everywhere, then why isn’t your AI?”

For real-time responsiveness, enterprises need to be able to put their AI everywhere too, including right next to where the decisions are being made.

But the reality is that centralised models alone don’t cut it for machine-scale speed in a distributed world. Enterprises need to rethink their architectures - not just for performance, but for resilience, agility, and autonomy.

4. The future is decentralised and agentic - humans will always be needed, but what will be our role?

The Age of AI isn’t a continuation of the internet age. Currently, it’s the opposite:  closed, centralised, and dominated by a handful of players.

As Ray warned, “You can’t operate in a decentralised world with the current models”.  The future is agentic, built on autonomous systems that make decisions at speed, closer to the edge, and in real time.  

Agentic AI isn’t just about automation. It’s about decision velocity – making faster, better, cheaper decisions at machine scale.

But this raises critical questions:

• Where and when do you put the human in the loop?

• Can you operate at machine scale with humans?

• Are we able to take accountability for autonomous system decisions?

Agentic systems will reshape industries – but they must be designed with humans in mind. For every technology evolution, successful adoption has depended on aligning people, processes and technology for a thoughtful purpose that considers wider societal, environmental and economic impacts.  

5. We must equip current and future generations to thrive in an AI-automated world

AI is more than just a technology shift; it’s a societal one.  

Professor Savvas Papagiannidis highlighted the urgent need to rethink education, skills, and workforce development: 

“Universities must evolve to equip the next generation with the skills needed to thrive in an AI-automated world. It’s not just about technical skills but raises broader social and ethical questions about how humans can thrive in this environment.”

At the end of the day, AI isn’t human. It lacks authenticity, empathy and the human connection needed to build trust in so many of our basic interactions.

As Ray noted, we’ll always need humans in the loop to “automate precision decisions, which require context, ethics, and accountability - things only humans can provide.”

The race is on: are you keeping pace?

The roundtable ended with a sense of urgency. The Age of AI is here. The rules have changed. And the winners will be those who move fast, think differently, and build for a decentralised, agentic future.

Global forces - from geopolitical shifts to technological disruptions - are accelerating the race to real time. Enterprises must respond with purpose with innovative, bold strategies and new architectures.

Plan for decentralised AI. Connect the dots. Apply industry expertise with new techniques. Assume machine scale - but build in humans.

Author: Danielle Honoré, Head of Marketing and Alliances