Amidst the fervor surrounding AI’s transformative potential – consultants forecasting trillion-dollar opportunities, social commentators envisioning entirely new ways of living – there’s a growing sense of disillusionment. Leaders grapple with uncovering industry-changing benefits, while employees feel overwhelmed by the simultaneous demand to boost productivity and master a new set of skills.
I’ve engaged with dozens of leaders and observed numerous teams navigating AI adoption over the past 18 months. This has provided deep insights into the real challenges impeding the AI revolution. To understand them, let’s revisit a historical technological shift: the advent of electricity.
Before electricity, factories were designed around a large, central steam engine. This setup required a complex arrangement of shafts, belts, and pulleys to power various workstations, resulting in box-shaped factories constrained by the limitations of mechanical power transmission. When electricity emerged, many factories simply replaced the steam engine with a large electric motor. While this improved reliability, it didn’t address fundamental inefficiencies – the factory layout remained unchanged, still tethered to a central power source.
It wasn’t until a generation later that managers realized electricity’s true potential. By installing smaller electric motors at individual workstations, factory layouts could be completely redesigned. This shift ultimately led to the modern production line, marking a significant leap in productivity and giving rise to the factory as we know it today.
“To unlock the real power of AI, we need to frame it in terms of solving fundamental challenges rather than merely enhancing efficiency. ”
Much like the early days of electricity, AI is often implemented today as a direct substitute for existing tools, without rethinking underlying processes. Organizations adopt AI to perform tasks faster – creating presentations, writing text, generating images – but rarely consider how it could fundamentally transform their business models.
One core issue is the challenge of measurement. AI impacts knowledge work, an area notoriously difficult to quantify. This lack of clear metrics contributes to vagueness and uncertainty. Managers may demand productivity increases through AI without a concrete understanding of “how” or “how much.” Moreover, when an employee’s work isn’t measured, their motivation to embrace change diminishes. Faced with the choice between producing something 30% “better” – an improvement that’s not quantified – or avoiding the effort of learning a new skill, many opt for the latter. After all, most employees aren’t evaluated or rewarded based on creating “better” work, but on meeting a broad range of intangible management expectations. For new technologies to take root (like, for example, email did), they must offer tangible benefits beyond the general concept of “better”.
Take Microsoft Copilot for example. Paying $30 per employee to provide an excellent productivity solution for your organization is one thing, but convincing people to adopt a new way of working that maximizes this investment is another challenge altogether. What we should remember is that most of the value created by computers historically was not in increasing efficiency, but by completely replacing functions such as expensive manual bookkeeping that no longer needed to be done.
To unlock the real power of AI, we need to frame it in terms of solving fundamental challenges rather than merely enhancing efficiency. This requires a shift from incremental improvements to transformative thinking. AI applications need to make the previously impossible become possible.
A recent example comes from the early days of the internet. In the late 1990s, companies treated online advertising like print, placing static banner ads on websites. These ads were not hugely effective, leading to the saying, “trading analog dollars for digital cents.” The game changed when Google introduced a performance-driven model with Google Ads, allowing advertisers to pay only when users clicked on their ads. Facebook further revolutionized media consumption with the News Feed, personalizing content delivery.
These innovations didn’t just digitize existing models: they leveraged technology to create new business paradigms impossible in the pre-digital era. Today, Google and Meta own the majority of the $300bn digital ads industry because they reimagined media in a way that was not possible pre-internet.
Currently, we are at the “digital banner ad” (or large electric motor) stage with generative AI. Despite extensive discussion of AI’s potential to transform our lives, most applications enable us to perform existing tasks faster rather than to create new value exponentially. The next generation of AI solutions must unlock opportunities that were unattainable before AI.
In a scenario where managers are inundated with AI narratives – often leading to confusion – it’s crucial to develop a consistent and persistent message.
One obstacle to system transformation is our tendency to think linearly, making it difficult to grasp exponential trends. This mindset poses a challenge when introducing groundbreaking technologies. Balancing immediate needs with transformational, long-term benefits without overwhelming stakeholders is a delicate task as transformational shifts won’t yield instant returns, and focusing on immediate gains fails to convey the full story.
In a scenario where managers are inundated with AI narratives – often leading to confusion – it’s crucial to develop a consistent and persistent message. Viewing AI as a journey and a long-term strategy, rather than a quick fix for annual targets, is essential to exploit its true potential.
This begs a pivotal question: Where in the value chain should AI-driven transformation begin? Initial speculation suggested that AI might disintermediate agencies integral to the communications and marketing ecosystem. However, this overlooks the vital role agencies and consultancies play in driving change and buffering organizations from the disruptive effects of transformation.
At InferenceCloud we have embraced this reality by partnering with agencies to “AI-enable” this critical segment of the value chain. For brands, this means accessing a new breed of AI-powered agencies capable of delivering superior results without overhauling internal processes. For agencies, it offers a path to differentiation and exponential growth. Traditionally, agency costs and revenues scale in a linear way. By integrating AI, agencies can scale operations exponentially.
The key to harnessing AI’s full potential lies in relentless innovation in all of these respects. The AI-powered economy will differ vastly from today’s. Commoditizing intelligence presents enormous economic benefits for those adept at leveraging it, but it demands innovation across all facets – core technology, application development, change management, and business models.
This transformation is far more profound than deploying chatbots for customer service or automating meeting schedules. It calls for visionary thinking and a willingness to reimagine foundational aspects of business.
The journey to unlocking AI’s transformative power is complex and multifaceted. It demands that we move beyond surface-level implementations and question the very foundations of our business models. By learning from past technological shifts and embracing innovative thinking, we can harness AI to create unprecedented value. The path forward may not be straightforward, but for those willing to lead the way the opportunities are boundless.
Reinventing today’s economy with AI is going to take time, but our approach to innovation is rooted in solving previously intractable problems with technology. Here are three examples:
Media clippings and social dashboards tell you what’s happening, but do they give you actual insight? They are also time-consuming to prepare and to read. We asked ourselves: “What is the core problem that media monitoring should solve?”
Influence Mapper answers this using AI to analyze inside-out brand communication and outside-in media perception to understand the core messages and perceptions, enabling communicators to quickly understand the big picture.
We then compare these messages against millions of external data points to predict effectiveness and impacts and formulate an appropriate strategy.
The result of this process is that, rather than a simple list of clippings or a social dashboard, we can provide deep insights and concrete recommendations.
A huge amount of time is invested in building, sharing, presenting, and discussing communications analytics dashboards. But how many decisions are made as a result, and how do these dashboards drive improvement? Such data can give you a good picture of the status quo, but it cannot predict improvement and recommend the next actions.
By building a unique dataset based on millions of online conversations, we can accurately predict the topics and messages that will most resonate with a specific audience to achieve a specific communications objective. We use this data to provide concrete strategy recommendations. This approach makes analytics faster and more meaningful and helps drive improvement.
Communication planning is a time-consuming business; particularly with the need to consider multiple audiences, geographies, and data. And, in a fast-moving world, the need to constantly update plans compounds this challenge. We asked ourselves: “How could AI use data to optimize and automate this process?”Â
InferenceCloud Strategy Builder uses our conversational graph technology to understand the gap between the current position of an audience and the communications objective. By plotting the optimum path across this gap, we are able to build systematic plans that convey a compelling narrative to the audience. Â
By continually monitoring audience feedback via analytics, plans can be dynamically updated to ensure they are directly meeting audience needs. This approach saves considerable time and makes sure that communications plans are precisely optimized for the audience.Â
IMD’s Simon J Evenett, Johannes Fritz, and Tommaso Giardini argue that the Trump administration will take a more sophisticated approach to AI regulation than some expect ...
GenAI is becoming a more common feature of the supply chain function, but significant areas of opportunity remain, suggests IMD’s Carlos Cordon. ...
AI is rewriting the rules of hiring – job seekers and recruiters alike must adapt to a process driven by algorithms but grounded in human connections. ...
Breakthroughs in agentic AI, open-source development, and closer collaboration between AI companies and the defense sector have significantly raised the AI risk profile. These developments have compelled us to advance the AI...
Explore first person business intelligence from top minds curated for a global executive audience