DAVOS, Switzerland — This week, I engaged in discussions with technology leaders at the World Economic Forum in Davos regarding their goals for the upcoming year and the major trends they anticipate influencing investor attitudes.
Similar to last year, AI was the focal point of the conference. However, this year’s conversations shifted from AI algorithms or which chatbot excels, to the strategies enterprises will use to integrate this technology and the advancements that lie ahead.
These factors are likely to be critical in determining the revenues and stock values of the world’s leading tech enterprises.
Adoption of Enterprise AI
This year will see a rise in AI adoption across companies, but the methodology will be distinctly different from 2025. Many organizations experimented with AI pilot programs last year that did not transition into full-scale implementation.
Previously, organizations embraced AI tools, such as chatbots from Microsoft, OpenAI, and others, driven by the fear of being left behind. Dowson Tong, the CEO of Tencent‘s cloud division expressed in a recent CNBC interview.
This time, organizations will take a more discerning approach to the types of AI they implement.
“This year, in discussions with clients, I believe they have become much more practical. They seem to have moved beyond the phase of FOMO,” Tong expressed. “Today, many clients are significantly more precise.”
Raj Sharma, global managing partner for growth and innovation at EY, indicated that businesses must rethink entire processes that could utilize AI. “That’s when the true value will start to emerge,” Sharma remarked in a CNBC interview.
Agentic AI
This was among the most talked-about terms of 2025 and is likely to remain a prominent topic. Agentic AI generally refers to AI systems capable of performing tasks on behalf of users. The ultimate aim is to develop intricate agents that operate independently with minimal human engagement.
Technology chiefs informed me that while agents are being deployed, the scale and capabilities vary widely. It also heavily depends on the specific industry in question. Some uses of agents in enterprises involve executing straightforward processes, while others engage in more complex workflows.
“I believe these agents are not self-governing. I don’t think we’ve reached the stage where human employees can be replaced,” stated Uljan Sharka, CEO of the startup Domyn.
The most optimistic about the technology was Fabricio Bloisi, CEO of Prosus. He mentioned that Prosus is currently operating 30,000 agents, and within the next five years, there could be companies managed by agents.
“I don’t think it’s merely hype and optimism,” Bloisi stated during an interview with CNBC’s Dan Murphy and myself this week.
Geopolitics
The influence of geopolitical instability was a significant topic in many of my discussions this week.
“One aspect that we are not addressing in the AI arena, yet which will emerge as a major factor, either as a deterrent or an enabler, are the [geopolitical] challenges,” indicated EY’s Sharma.
China’s technological progress in sectors like AI and semiconductors was also among the conversation points. In a recent interview for CNBC’s new podcast, The Tech Downland, conducted with Steve Kovach, Google DeepMind CEO Demis Hassabis suggested that China’s AI models are merely months behind those of the U.S. and the West.
Physical AI
This year, physical AI is expected to feature prominently in tech discussions. The term refers to implementations where AI assumes a physical presence, encompassing everything from robotics to autonomous vehicles. My own experiences at Davos underscored just how tangible this movement has become: one evening, a robot was seated right at the dining table.
EY’s Sharma described physical AI as the “next wave,” estimating it could potentially be five to six times the market size of agentic AI in the next five to six years.
Conversely, Sassine Ghazi, CEO of semiconductor design tools firm Synopsys, noted that he initially anticipated physical AI would arrive “five plus” years from now, but it is emerging “much sooner.”
Jensen Huang, CEO of Nvidia, remarked that AI robotics represents a “once-in-a-lifetime” opportunity for Europe, given the region’s robust industrial manufacturing foundation.
