Davos 2025: AI Agents Shift From Hype to Business Reality

As business and tech leaders gather at the World Economic Forum in Davos, the conversation around artificial intelligence is taking a decisive turn from hype to practical implementation. The annual Swiss mountain resort gathering, which begins in earnest on Tuesday, is seeing AI agents emerge as a dominant theme alongside discussions of Donald Trump’s inauguration, tariffs, and cryptocurrency regulation.

Mihir Shukla, CEO of Automation Anywhere, told Business Insider that 2025 marks a pivotal shift in how executives approach AI. “As CEOs here are chasing AI, they find it cool, but the job is to find value,” Shukla explained. “I know this is cool, but what can I do with it? I think that conversation happens this year.” His company has already demonstrated the practical power of AI agents by building a boardroom AI agent trained on years of financial data and presentations. The agent proved so effective at pattern matching—surpassing even the most experienced human analysts—that the company gave it an empty seat at the boardroom table for dramatic effect.

The shift toward practical AI implementation is forcing major companies to rethink their entire business models. Raj Sharma, EY’s global managing partner for growth and innovation, revealed that the professional services giant is reconsidering its commercial approach. Instead of traditional hourly billing, EY is exploring a “service-as-a-software” model where clients pay based on outcomes rather than time spent. This transformation raises critical questions about ownership, accountability, and revenue attribution when AI agents perform the work.

“As much time as we are spending on the technology aspect of these things, the whole commercial model, the risk models associated with that, there’s an equal amount of energy, at least companies like ours are spending, to say ‘What is the right commercial model?’” Sharma noted. The challenge extends beyond external client relationships to internal structures—companies must determine who owns the agents and takes responsibility for the revenue they generate.

This year’s Davos represents a three-way battle for attention between agentic AI, Trump’s tariff policies, and cryptocurrency regulation. The presence of emerging economies like Brazil, Indonesia, Mongolia, and Korea—joining previous years’ participants India and Saudi Arabia—adds another dimension to discussions about how AI and other technologies will shape global economic development.

Key Quotes

As CEOs here are chasing AI, they find it cool, but the job is to find value. I know this is cool, but what can I do with it? I think that conversation happens this year.

Mihir Shukla, CEO of Automation Anywhere, describes the fundamental shift happening at Davos 2025, where business leaders are moving beyond AI enthusiasm to focus on practical value creation and measurable business outcomes.

It was able to match patterns that the most experienced people couldn’t.

Shukla explains the capabilities of his company’s boardroom AI agent, which was trained on years of financial information and presentations. The agent’s performance was so impressive that the company gave it a symbolic seat at the boardroom table.

As much time as we are spending on the technology aspect of these things, the whole commercial model, the risk models associated with that, there’s an equal amount of energy, at least companies like ours are spending, to say ‘What is the right commercial model?’

Raj Sharma, EY’s global managing partner for growth and innovation, highlights how AI agents are forcing professional services firms to fundamentally rethink how they charge clients and attribute revenue internally, moving from hourly billing to outcome-based pricing.

Our Take

The Davos 2025 discussions reveal that the AI industry is entering its most consequential phase yet. The transition from “what can AI do?” to “how do we capture value from AI?” marks genuine maturity. What’s particularly striking is how AI agents are forcing companies to confront uncomfortable questions about their fundamental business models—questions many have avoided for decades. EY’s shift toward outcome-based pricing could trigger a cascade effect across professional services, consulting, and knowledge work industries. The boardroom AI agent example is both exciting and sobering: if AI can outperform experienced analysts at pattern recognition, we’re not just automating tasks but potentially replacing entire categories of high-value work. The real test will be whether companies can navigate the organizational politics of AI ownership and revenue attribution—technical challenges may prove easier to solve than human ones.

Why This Matters

This shift from AI hype to practical implementation represents a critical maturation point for the artificial intelligence industry. After years of breathless excitement about AI’s potential, business leaders are now grappling with fundamental questions about value creation, commercial models, and organizational structure. The fact that major professional services firms like EY are reconsidering decades-old billing practices demonstrates how profoundly AI agents could reshape entire industries.

The boardroom AI agent example from Automation Anywhere illustrates that AI is moving beyond experimental use cases to core business functions like financial analysis and strategic planning. When AI can outperform experienced analysts at pattern recognition, it raises urgent questions about workforce transformation, skill requirements, and the future of knowledge work. The commercial model challenges highlighted by EY signal that companies must simultaneously innovate on technology and business strategy—those who fail to adapt their revenue models may struggle to capture AI’s value even if they successfully deploy the technology. For emerging economies represented at Davos, these AI developments could either accelerate development or widen existing gaps, making international coordination increasingly important.

For those interested in learning more about artificial intelligence, machine learning, and effective AI communication, here are some excellent resources:

Source: https://www.businessinsider.com/davos-world-economic-forum-trump-inauguration-ai-day-one-2025-1