Retorio Blog

From hype to value creation: How AI coaching transforms human performance at scale

Written by Dr. Patrick Oehler | 01.10.2025

This article reflects my key takeaways from Retorio’s 2025 Tech 'n' Talks Wiesn event, enriched by the inspiring contributions of Prof. Christian Gärtner (Munich University of Applied Sciences), Rolf Fricker (Oliver Wyman), Chetak Buaria (Merck Group), Marc Städing (Catacumbo), Armin Lutz (BWV Nordbayern Tühringen), Sebastian Hahn (Die Bayerische), Robert Jung (Nürnberger Versicherung), Alessandra Tieghi-Krakowiak (Baxter), Evren Karayel (Amgen), Christoph Kunz (rpc) and many more.

The AI paradox: Adoption without impact

The global race to adopt AI is moving at full speed. Today, 8 out of 10 companies claim to use AI. And yet, according to McKinsey’s 2025 report, 80% of organizations fail to see measurable business results. Even more striking: AI has not yet demonstrated a consistent effect on profitability.

This raises the central question: how do we turn AI hype into tangible value?

Why business impact lags behind

Prof. Christian Gärtner outlined several reasons why AI’s promise often fails to translate into bottom-line results:

1. The Red Queen hypothesis

In Lewis Carroll’s Through the Looking-Glass, the Red Queen says: “It takes all the running you can do, to keep in the same place.” The same applies to AI. Simply adopting AI isn’t enough — companies must continuously evolve and improve to keep pace. To actually advance, they must run twice as fast as competitors.

2. The trap of delusional AI output

The realism of GPT-based systems can be misleading:

  • Hallucinations = Botshit: AI fabricates convincing but false information.

  • Sycophancy: Bots flatter and agree with users, reinforcing bias.

  • Realistic Interaction: Chatting with AI feels like talking to a trusted friend, blurring the line between machine and empathy.

  • Cognitive Dissonance: Users know AI is “just code,” yet they respond as if it were intelligent—sometimes creating confusion or paranoia.

These psychological traps can derail productive AI use if not managed carefully.

3. Perception vs. Reality of work

Job descriptions make work look linear and simple. Reality is far messier: employees interact with colleagues, customers, external networks, cloud systems, vendor platforms, and expert support.

If AI is trained only on simplified models (like a single system), it misses the full complexity of how work actually happens. The result? Incomplete automation and ineffective coaching.

4. The democratization of performance takes time

Perhaps the most exciting insight: AI helps novices more than experts, yet experts often adopt it first.

  • Experts gain a small productivity boost—helpful, but incremental.

  • Novices experience a leap in performance—AI closes knowledge gaps, handles basic tasks, and levels the playing field.

Gen AI can lead to a democratization of performance but often fails to do so (Gärtner, 2025) 

In other words, AI can democratize performance when novices adopt it at scale. Its true value lies not in making the best slightly better, but in raising the overall floor of performance across organizations. 

From promise to practice: How to unlock AI’s value

How can companies break through the paradox and realize measurable business outcomes? Here are the key action items:

  • Solve the “unsolvable.” Don’t limit AI to low-stakes tasks. Test its power on problems once considered too complex or resource-intensive.

  • Start with the business case. Always ask: So what? AI should address core business problems, not serve as a shiny toy.

  • Prepare for DIY. Many repetitive tasks can shift to employee self-service through GenAI—if governance and safeguards are in place.

  • Balance exploitation and exploration. Use AI both for structured efficiency (accuracy, compliance) and for experimentation (creativity, innovation).

  • Understand AI’s limits. Technology alone won’t deliver results. Success requires the right data, mindset, skillsets, incentives, and organizational structures.

As Prof. Gärtner put it: “IA before AI.” Solid information architecture and compliance (e.g., under the EU AI Act) must come first. Only then can companies scale AI beyond isolated pilots into portfolios of initiatives tied to real outcomes.

Industry insights & best practices

This year’s Tech ’n’ Talks Wiesn 2025, featured leaders from pharma, insurance, and automotive who discussed AI (coaching) use cases in different industry-focused streams, yet exploring one question: how can AI unlock commercial excellence at scale?

The discussions revealed both opportunities and roadblocks. Regulatory hurdles, political friction, and the rising skepticism around omnichannel effectiveness continue to challenge adoption. Meanwhile, patients are turning to ChatGPT for medical advice, and healthcare professionals are inundated with automated sales outreach.

In this environment, one trend became clear: we are witnessing a renaissance of the human sales rep. Trust, credibility, and authentic relationships remain decisive. And when empowered with AI coaching, sales professionals can amplify their impact while keeping humans at the center of the value chain.

Pharma & Life Sciences: Humans first, tech second

Merck Group, recently ranked #2 globally in the AI Readiness Index, highlighted a crucial principle: success in AI and AI coaching doesn’t begin with technology. As Chetak Buaria, VP of Global Commercial Operations Oncology, stressed: “People, then process, then technology.” Iteration, lean pilots, and rapid scaling are vital—but only when humans remain the anchor.

Marco Rauland, VP Global Market Access and Pricing at Merck group sees widespread experimentation, often with limited impact: Everyone is working intensively with AI, but most use cases are still trial-and-error with modest effect on top and bottom line, in part due to hallucinations that prevent promising use cases, such as dossier creation. He sees comparably stronger results for AI coaching, as they present a valuable opportunity to practice negotiations with payers. He sees more and more applications beyond pharma companies: Payors and AMNOG stakeholders are also applying AI, assessing the quality of value dossiers but facing similar challenges as big pharma.


Pharma AI Readiness Index (CB Insights, 2025)

Baxter echoed this, emphasizing compliance as the foundation. Their top-down strategy ensures employees are trained on regulatory standards before exploring AI tools. Yet, their challenge is equally human: how do you create space for employees already drowning in back-to-back meetings to learn, experiment, and apply new capabilities?

Amgen urged the audience to separate hype from reality—especially in high-stakes areas like oncology. They coined a striking phrase: “role play without role play.” AI coaching allows reps to practice objections, refine their messaging, and build confidence in a safe environment—before stepping in front of real physicians.

Insurance: Bridging generations with AI

The German insurance sector faces a very different challenge: demographic change. Companies such as Allianz, VKB, NÜRNBERGER Versicherung, die Bayerische, and Helvetia are rethinking product design, customer engagement, and sales enablement.

Here, knowledge transfer across generations is mission-critical. Baby Boomers and Gen X hold decades of tacit expertise, while younger employees demand digital-first, fast learning. The risk? Valuable expertise could disappear with retirement.

AI coaching offers a bridge:

  • Preserving and structuring expert knowledge.

  • Scaling mentorship across teams.

  • Personalizing development for every generation—from Gen Z’s hunger for instant feedback to senior experts’ need for knowledge validation.

The conclusion: no single generation is “the challenge.” Each brings unique strengths and gaps. What matters is building a system that combines experience with fresh perspectives—something AI coaching is uniquely positioned to enable.

Automotive & Retail: Customer-driven, Not Hype-Driven

The idea that “AI will replace humans” in Automotive & Retail oversimplifies reality. What AI truly replaces are tasks, not people. This makes it even more critical to define where humans add irreplaceable value.

The essential human skills in the AI age are:

  • Critical thinking: spotting hallucinations, validating accuracy, exercising judgment.

  • Emotional intelligence: building authentic, trust-based customer relationships.

  • Creativity & adaptability: innovating where predictive logic falls short.

Luxury brands like BMW and Porsche are experimenting with AI-driven sales training to improve customer satisfaction. According to Christoph Kunz, Founding Partner and President of RPC (Retail Performance Company, a joint venture of BMW Group & H&Z), the success factors are clear:

  • Projects succeed when they’re customer-driven, iterative, and lean.

  • Projects fail when they chase hype without grounding in human behavior.

Once again, the human factor is decisive: AI succeeds only when it enhances—not replaces—authentic customer interactions.

A Common Thread: AI as a Human Amplifier

Across industries, one message rang loud and clear at Tech ’n’ Talks Wiesn 2025: AI is not the star of the show—humans are.

Whether in pharma, insurance, or automotive, commercial excellence depends on trust, credibility, and adaptability. AI’s role is to amplify human potential—to provide a safe space where sales reps, advisors, and consultants can practice, learn, and scale their performance safely.

As Chetak Buaria (Merck Group) puts it:  AI doesn't replace human skills – it helps them unlock their full potential. He describes a battle of two AIs.

Actual Intelligence (Humans):
  • Humans are positioned to focus on ideation and innovative thinking.
  • The role of the human is to collaborate on complex decisions.
  • AI allows humans to outsource scalable tasks, freeing up cognitive resources for higher-level work.

Artificial Intelligence (AI):
  • AI excels at navigating the vast digital ocean.
  • Its primary function is to help humans find, translate, and synthesize pre-existing information and data.

In his view, the "battle" is not a competition but a division of labor where humans manage the creative, strategic, and decision-making aspects, while AI handles the heavy lifting of data processing and synthesis. The future of commercial excellence isn’t about replacing humans with technology. It’s about ensuring humans remain at the center—empowered, trusted, and equipped to make the decisive difference.

Conclusion: From Hype to Human-Centric Value

The story of AI in 2025 is no longer about who adopts it fastest—it’s about who translates adoption into sustainable value.

Generative AI significantly boosts novice performance (over 40% gain) but offers only marginal improvements for experts. This highlights a crucial paradox: as AI advances, the "human edge" becomes increasingly vital. Our unique contribution isn't speed, but purpose – why we work, which demands a redefinition of human potential.

Leaders who will flourish in this new era will:

  • Prioritize Information Architecture (IA) over Artificial Intelligence (AI): Establish robust foundations in data, governance, and compliance.
  • Invest in Portfolios, Not Pilots: Scale only solutions that demonstrate proven value.
  • Strengthen Human Differentiators: Emphasize critical thinking, emotional intelligence, and adaptability
  • View AI as a coach and collaborator. Leverage AI for its strengths—data analysis and pattern identification—to free humans for uniquely human tasks like creativity, empathy, judgment, and relationship-building. 

AI won’t magically transform organizations. But when combined with the right structures, mindsets, and human skills, it can democratize performance, unlock creativity, and prepare employees to thrive in complex, human-centered work.