AI’s Reality Check

Why Overpromising Breaks More Than It Builds

When generative AI surged into mainstream consciousness in late 2023, marketing and creative agencies immediately jumped in with the belief that AI could supercharge productivity, enhance creative output, and elevate efficiency. Yet a recent Orgvue survey highlights a stark reality check: 55% of executives admit disappointment, recognizing that AI-driven gains haven't arrived as rapidly or substantially as expected. (The Guardian, May 2025.)

This realization isn't a rejection of AI. Rather, it's a call to recalibrate expectations, acknowledging that the industry was oversold an overly simplified vision of AI’s potential. The real costs of this misunderstanding extend beyond lost productivity, they ripple through company culture, trust, and innovation capacity.

The Overpromise Trap

Generative AI arrived with irresistible appeal: a seemingly revolutionary tool capable of generating compelling copy, striking visuals, insightful strategies, and hyper-personalized experiences instantly and at scale. It felt like magic. Agencies, already facing intense pressure to demonstrate rapid, tangible results, saw AI as a productivity booster, potentially removing creative bottlenecks and amplifying human efforts overnight.

Yet efficiency without fluency leads to operational friction. Many organizations prematurely invested in AI solutions expecting immediate returns, only to encounter limitations. Rather than streamline workflows, hastily integrated AI often introduced new complexities and uncertainty.

AI's current state is inherently assistive rather than autonomous. It can augment creative efforts, spark ideas, and streamline repetitive tasks. But entirely replacing complex roles and workflows without significant human oversight and detailed process redesign is unrealistic.

Recognizing AI’s True Maturity

Top-tier generative AI is powerful yet fundamentally limited. It excels in tasks like content synthesis, initial research, or basic creative ideation, but struggles significantly with nuanced brand identity, tone consistency, strategic context, and creative originality. All the “agentic” automation hype is overblown, for now.

Leaders in the field openly acknowledge this gap. For example, WPP developed "WPP Open," an AI-driven creative platform designed to generate campaign content. Yet, according to WPP’s CTO, the platform's value emerges clearly only when paired with human creativity and strategic oversight. AI supports the creative process but doesn't define it. (The Times. May, 2025.)

Additionally, integrating AI effectively requires robust new skill sets within teams, specifically around validating, auditing, and refining AI outputs. Without these skills, AI adoption becomes a hindrance rather than an asset, adding friction rather than efficiency.

The challenge isn't technology itself, but the simplistic assumption that adopting AI tools alone transforms business outcomes. Effective integration demands reimagined operating models otherwise, AI scales problems instead of solutions.

The Real Cultural Cost

While sweeping layoffs directly triggered by AI haven't widely occurred within agencies, premature promises and vague integration strategies have inflicted significant cultural damage. Leadership enthusiasm unintentionally sowed confusion and anxiety, creating uncertainty around the future roles of creative teams.

Employees felt uneasy, questioning their value when AI was championed without clear frameworks for human-AI collaboration. Even without immediate layoffs, the lingering threat of obsolescence eroded morale and stifled creativity, undermining trust and diminishing creative confidence.

Rather than causing direct, widespread job cuts, poorly executed AI integration risks long-term cultural damage, loss of institutional knowledge, erosion of creative chemistry, and diminished innovation capacity. Trust in leadership deteriorates when bold promises remain unfulfilled.

What Balanced Integration Actually Looks Like

The solution isn’t rapid automation but thoughtful integration. Silicon Valley might champion speed and scale, but sustainable AI integration prioritizes thoughtful enhancement of human capabilities over automation. Agencies should focus first on how AI can reduce friction, accelerate mundane tasks, and free creative teams for strategic, high-value activities.

Rather than broadly asking, "What can AI automate?" leaders should examine, "How can AI best augment our people?" This reframing leads to intentional reallocation of tasks rather than wholesale replacement. Agencies need to develop clear training programs, refined workflows, and new methodologies explicitly designed for seamless human-AI collaboration.

The goal should be thoughtful reallocation rather than wholesale replacement. AI should function as an amplifier for human creativity and strategy, accelerating workflows and enhancing outputs without threatening roles.

To achieve this, businesses need clear training, revised workflows, and refined methodologies that incorporate AI as a seamless extension of human work. Essential processes — from brief development to feedback loops, quality assurance, and delivery — should explicitly delineate human-AI collaboration. The “how” of this is tricky and nuanced. It’s something I’d like to dive into a bit more in coming posts, so, more to come.

Leadership Imperatives for Sustainable Integration

A reality check isn't a setback, it's vital recalibration. Early-stage technological adoption inevitably encounters hype, missteps, and adjustments. Shockingly to most enthusiasts, AI is not different. Humans drive real change still, not the tools themselves. Successful leadership involves learning and adapting rather than doubling down on failed assumptions.

Operational clarity must become paramount. Leaders must critically examine AI integration thus far: Where did it succeed? Where did it fall short? What precisely do teams need to integrate AI effectively?

Cultural restoration is equally important. Transparent communication about AI’s capabilities and limitations helps rebuild trust and momentum. Teams should see experimentation as a growth opportunity, not a risk. Encouraging a culture where human creativity remains central, augmented by AI tools, fosters genuine innovation.

Ultimately, successful integration requires leadership humility: learning instead of layoffs, patience over speed, and a disciplined approach instead of quick fixes.

Making the Better Bet

The promise of AI remains profound, but true advantage lies in thoughtful integration rather than rapid adoption. Businesses don't simply need new AI tools, they require fundamentally new ways of working, measuring value, and fostering collaboration. That’s something I wish to focus more on in the coming weeks and months because it’s the thinking with these tools that is really hanging up leaders and first-movers.

It may be less of a sexy thought, but it really is a leadership and strategic problem. Not a tool problem. Getting AI right isn't about moving fastest, it's about moving thoughtfully, ensuring AI genuinely enhances the value humans uniquely provide.