The World Economic Forum (WEF) in Davos this year offered a mix of inspiration and grounded reflection for attendees. One thing that stood out immediately was the weather—warmer than previous years, a subtle yet stark reminder of climate change’s presence, even amidst discussions dominated by technology and innovation. Yet, while the temperatures outside hinted at environmental challenges, inside the halls and sessions of Davos, the temperature of excitement around Artificial Intelligence (AI) was hotter than ever.
The narrative from Davos was clear: it’s all AI, all the time. From keynote speeches to roundtable discussions, every corner of the forum echoed AI’s transformative potential. More than just hype, though, there was a palpable sense that we’re entering a pivotal phase—not just of adoption, but of understanding what AI truly means for businesses, governments, and society.
AI Agents: The Next Evolution
One recurring theme was the rise of AI agents. These are no longer theoretical concepts. They’re emerging as practical tools for streamlining workflows and addressing complex challenges. Several speakers emphasized how AI agents, powered by domain-specific large language models, are replacing repetitive, labor-intensive tasks. Marcel Stalder, CEO of Chain IQ, shared how his firm leverages AI agents to transform procurement workflows. “We identify labor-intensive activities within processes and deploy agents to replace them,” he explained. “It’s about domain-specific AI that provides real, measurable results.”
This vertical, purpose-driven AI is a critical next step, moving beyond generic capabilities toward tailored, high-impact solutions. Panelists consistently highlighted the importance of starting small. Rather than setting unrealistic goals for AI to replace the most complex human tasks, businesses are encouraged to focus on achievable, high-volume areas that yield measurable ROI. As Rahul Kanodia, CEO of Datamatics, advised, “Don’t try to make AI run before it learns to walk. Start with processes that deliver immediate value.”
“Don’t try to make AI run before it learns to walk. Start with processes that deliver immediate value.”
Optimism Amidst Uncertainty
Amidst the intense discussions, a surprising tone of optimism emerged about the year ahead. Industry leaders acknowledged the challenges of deploying AI responsibly, the complexities of scaling, and the high costs involved. Yet, there was a shared belief that collaboration and innovation could overcome these hurdles.
Kanodia shared a compelling example: his team helped a company reduce transaction processing times from eight minutes to four and cut staffing needs by more than half. “This isn’t just about productivity,” he noted. “It’s about unlocking new opportunities for growth and innovation.” Such stories exemplify why optimism is warranted—when executed well, AI can drive both profitability and societal benefit.
Several other speakers echoed this optimism, pointing to advances in personalized services and increased operational efficiency. A particularly resonant statement came from LSEG’s Ron Lefferts, who highlighted how “AI is no longer about experimentation; it’s about delivering scalable value in real-time.”
The Dark Side: Technology and Our Kids
No discussion about AI and digital technology would be complete without addressing its darker implications. Several sessions explored the impact of technology on our children, raising concerns about screen addiction, misinformation, and the erosion of critical thinking skills. Panelists addressed these worries by emphasizing the importance of embedding empathy and ethical awareness into how we teach and use technology. As one speaker noted, “If we don’t teach the next generation to balance technology with empathy, we risk losing our humanity.”
This led to broader reflections on how to upskill society at large. As AI becomes ubiquitous, the skills we value most are shifting. Creativity, empathy, and judgment—traits that are uniquely human—are becoming the bedrock of future workforce training. Marcel Stalder put it succinctly: “We’re not teaching people how to beat AI; we’re teaching them how to work alongside it.”
Broader Trends and Takeaways
Beyond AI’s immediate applications, broader trends emerged across the sessions:
- The Democratization of AI: Many experts emphasized making AI accessible. From internal AI tools that empower non-technical employees to external platforms designed for mass adoption, the focus is on putting AI in the hands of as many people as possible.
- Trust and Data Governance: The importance of high-quality, trustworthy data cannot be overstated. Discussions centered on ensuring data integrity, transparency, and security, particularly in heavily regulated industries like financial services. “Volume of data is good, but quality of data determines success,” noted one panelist. The sentiment was echoed across sessions, with leaders calling for shared industry standards.
- AI’s Role in Productivity: The potential for AI to drive productivity gains remains immense. However, the consensus was that AI is still in its toddler phase. By focusing on iterative improvements, businesses can achieve consistent, scalable results rather than chasing perfection prematurely. “Even a 30-40% improvement in productivity is a huge win,” Kanodia reminded attendees.
- Ethical AI: Responsible innovation was another cornerstone of the conversations. Leaders urged for clear principles to guide AI development, balancing speed with accountability. As Stalder eloquently stated, “Innovation and responsibility are not mutually exclusive.”
Call to Action
Davos 2025 offered a microcosm of what lies ahead: a future where the role of AI will be both pervasive and profound. The conversations underscored the need for businesses to act decisively yet thoughtfully. Whether it’s adopting small language models to reduce costs or creating ethical frameworks for AI deployment, the key is to be proactive without losing sight of the broader implications.
As I reflect on my time at WEF, I’m struck by both the urgency and complexity of the challenges ahead. Bringing this technology to fruition isn’t just about overcoming technical hurdles—it requires navigating ethical, societal, and practical considerations as well. Yet, amidst these challenges, the opportunities remain vast and inspiring. By fostering collaboration, prioritizing education, and embracing innovation responsibly, we can harness AI’s potential to create a future that benefits everyone.
For those of us in leadership roles, the path is clear: we must be both stewards and innovators, ensuring that AI’s transformative power is wielded wisely. If there’s one takeaway from Davos, it’s that the future isn’t just something to predict—it’s something we must shape together.
Michael Ellison is the President of Corporate Insight.