By Javier Surasky
Today’s
leading international forums—such as the World Economic Forum in Davos (WEF),
the G7, or the G20—have become informal spaces where global-scale political
decisions are articulated. Although they lack the capacity to establish binding
rules, the political priorities defined by the countries gathered at those
meetings give rise to concrete policies, order international narratives, and
set interpretive frameworks that shape governance worldwide.
The case of
the World Economic Forum in Davos (WEF, World Economic Forum) is particular: if
the G7 represents the elite of industrial countries—contributing to the
convergence of their policies and making decisions that directly affect other
States—the WEF is an informal meeting of political, economic, and technological
elites that allows us to see the state of their relationships, identify
hegemonic narratives, and learn which discussions are open within the group.
The participation of leaders from countries of the South does not change that
situation; rather, it reinforces it: we hear their positions vis-à-vis the
agenda set by the elites and their demands toward those elites.
With that
understanding, we are interested in engaging with the discussions that the 2026
Davos meeting held on matters related to Artificial Intelligence (AI) because,
when a technology reaches high levels of diffusion and cross-cutting
penetration in social life, it ceases to be perceived as an innovation and
acquires a systemic character. The empirical indicator of that shift is its
repeated appearance in debates, in both the private and public sectors, which
stop being limited to technical issues and become integrated into economic,
social, and political discussions. This is exactly what has happened with AI.
This does
not mean that innovation within the technological field ceases; rather, work
moves to two different and intercommunicated planes: the plane of experts who
continue producing advances within the reference technology, and the properly
political plane related to its deployment, uses, and controls.
The 2026
Davos meeting crystallized that reality—which was in the process of
consolidation—within the plane of global elites, serving as an axis around
which debate was articulated. Not only was there discussion, especially from
the presence of the private sector, of the current situation and short- and
medium-term prospects for AI, but the countries present themselves included it
when presenting their growth expectations, their positions regarding existing
systemic risks, its impacts in economic and social areas, and—what may be
decisive—its condition as strategic infrastructure and as a constitutive
element of power.
On these
issues, divergent versions were laid bare—something characteristic of the Davos
Forum in times of international crises—since that space enables the coexistence
of competing narratives among representatives of the elites, under conditions
of access and recognition that are by no means democratic.
Starting
from these premises, we analyze critical elements that emerged from the recent
2026 Forum meeting and identify “constellations of positions,” a concept that
allows us to work with greater freedom and that we understand describes the
situation better than that of rigid “blocks.” Constellations of positions, or
simply constellations, represent contingent alignments, with possible overlaps
among actors investing different power resources in more than one constellation
depending on their interests, contexts, and objectives.
Speaking of
constellations also allows us to highlight the fact that we are not talking
about groups in which there is agreement on positions, but rather recurrent
patterns of argumentation integrated into a stratified space in which public
and private actors participate.
The WEF,
due to its high density of elites, allows us to conduct a single-case analysis
using a corpus that combines official Forum documents with high-visibility
interventions, applying an interpretive approach oriented toward identifying
problem definitions, frames of justification, and policy orientations. We do
not seek to infer causal mechanisms, nor do we start from the idea that the WEF
has global representativeness or legitimacy; rather, we want to build
mechanisms of framing and discursive legitimation that could influence how
certain courses of action become more or less possible in international arenas,
which ends up directly affecting opportunities for cooperation and conflict at
the global level.
Davos: a global elite’s stage
The WEF
Annual Meeting, held at the beginning of each year in Davos, is an informal
space for strategic deliberation among global elites. It does not adopt
decisions, but it operates as a platform for visibility and discursive
coordination among actors with a high capacity to build international agendas.
More than deciding, it helps stabilize priorities and interpretive frameworks
that are later contested in formal arenas (WEF, 2026).
The 2026
meeting took place between January 16 and 23 under the motto “A spirit of
dialogue,” framed in a context of persistent geopolitical tensions, economic
rivalry, a crisis of multilateralism, and technological acceleration. The
Global Risks Report 2026, prepared by the WEF itself before each of its
meetings, maintained that “uncertainty is the defining theme of the global
risks outlook in 2026 [that] marks an age of competition”.
That report
presents each year a ranking of perceptions of global risks, and in 2026 it
highlighted that in a “a contested multipolar landscape [in which] geoeconomic
confrontation has emerged as the top global risk for the short-term outlook,
ranking first in severity over the next two years” and “adverse outcomes of AI
is the risk with the largest rise in ranking over time, moving from #30 on the
two-year outlook to #5 on the ten-year Outlook” (WEF, 2026a:9), which clearly
reflects that concern grows as this technology is expected to reach new levels
of achievement and expansion.
Other
documents presented in Davos reinforce this framing and establish its links
with specific sectors: workforce skills adjustment, sectoral gaps in digital
transformation, energy consumption, infrastructure, and investment (WEF, 2026b;
WEF, 2026c; WEF, 2026d; WEF, 2026e). Along a complementary line, the report Planetary
Intelligence: AI for Global Risks and Resilience, also published by the WEF
(2026f), introduces the notion of “planetary intelligence” as an emerging idea
that combines AI with global observation and analysis capabilities to address
systemic risks, reinforcing its role as cognitive infrastructure.
In sum, in
Davos, the discussion about AI did not occur in a technical silo; rather, it
was inscribed in a holistic framework of competition and renegotiation of
leadership. To understand it better, we organize our analysis on two
intertwined planes: the technological and the geopolitical.
The technological approach
Here, the
leading role lay with the private sector, whether in companies directly linked
to AI development or in the representation of large investment funds.
Tensions
between accelerated innovation and social protection were recurrent, and two
phenomena struck us as especially noteworthy.
First, the
reframing of debates under the general umbrella of “social needs”: Is it better
to have, quickly, an AI that helps us solve urgent problems, such as climate
change, or to ensure the protection of people’s rights vis-à-vis this
technology? This idea generates a new frame that shifts the debate toward the
acceptability of certain costs: if AI promises improvements in security,
health, or productivity, what price—in terms of rights—would be considered
acceptable to accelerate its deployment? The answer to this question will
directly affect how alternatives aimed at reconciling speed of adoption and ex
ante safeguards are considered (such as “rights-by-design” approaches, or
integrating accountability mechanisms into model design).
Jamie
Dimon, CEO of JPMorgan, warned that AI can move “too fast for society” and that
governments and companies should intervene jointly to manage the transition, in
particular through reskilling and accompanying measures (Collingridge &
Wearden, 2026). Satya Nadella, CEO of Microsoft, however, linked social
legitimation to achieving tangible results: the AI community will lose the
“social permission to actually take something like energy, which is a scarce
resource” if its intensive use does not translate into improvements in health,
education, public efficiency, and competitiveness (WEF, 2026g, 00.09.18).
In a
complementary approach, Mohamed Kande (2026), president of
PricewaterhouseCoopers, argued that adoption still does not generate sufficient
value for many companies, pointing to a decoupling between technological
expectations and implementation capacities within the productive sector.
Both
tensions—external legitimacy and internal capacity—are condensed in Jensen
Huang, founder and CEO of Nvidia, who defended substantial investments and
stressed that we are witnessing the “largest infrastructure buildout in human
history” (Huang, 2026, 00:12:01). The greater the demand for infrastructure,
the greater the energy consumption and the greater the need to build social
acceptance through a narrative of “human needs” compatible with the urgency of
deployment.
In this
view, the proposed narrative seeks to achieve social acceptance by framing AI
governance as an enabling condition rather than a framework for establishing
agreements.
The geopolitical approach
In the opening, the President of the Swiss Confederation, Guy Parmelin, positioned digital technology as a challenge that cuts across the economy and society, noting that 2025 was a year of geopolitical, economic, and digital upheavals that “permeate our daily lives” (Guy Parmelin, 2026). Based on positions expressed at the Forum by national leaders and private-sector leaders, we identify three constellations of positions, which share Parmelin’s assertion as a starting point and which we present below—something that requires four specific clarifications:
- Davos is an elites’ forum with unequal access and visibility, so the constellations are not representative of a broader global map.
- The constellations are built through discursive framing, not by actor type, and can group States, companies, and leaders when they share the same argumentative orientation.
- They are ideal-typical clusters, not mutually exclusive fields, and actors can partially integrate into more than one of them.
- We considered the convenience of using finer partitions, but they did not add relevant information.
With those
conditions stated, the constellations we refer to are:
a) Pragmatic
implementers: with a functionalist approach, they seek to deploy AI to
solve practical problems and adapt organizations and economies. Here we place
countries that prioritize adoption according to their tangible impact, such as
India, the United Arab Emirates, and Indonesia. Subianto (2026), the president
of the latter, highlighted educational digitalization; Jafar, from the United
Arab Emirates, underscored the articulation among government, companies, and
philanthropy to absorb initial risks and invest in infrastructure (Jafar,
2026).
Perhaps the
best private-sector representative in this group is Elon Musk, who took part in
the Forum despite having criticized it for years, considering it a space of
elites without legitimacy. Musk focused his intervention on AI and robotics
applications, which he described as “essential for solving poverty, demographic
decline, and long-term economic stagnation” (Digital Watch, 2026).
b) Drivers
of competitive technological development: focused on AI as a strategic
infrastructure that provides systemic advantages, they prioritize accelerating
investments and consolidating leadership.
Among the
countries in this group, the United States stands out clearly, alongside Saudi
Arabia and Israel. Trump (2026) emphasized his country’s leadership in AI and
linked it to his energy and industrial policy, stating that the United States
is “leading the world in AI by a lot” and that it plans to “more than double
the energy currently in the country, just to take care of the AI plants”.
The other
major reference point is China. Its vice premier, He Lifeng, questioned trade
wars and highlighted Chinese leadership in research capacities and
international patenting (Lifeng, 2026).
In the
private sector, Nvidia stands out here, alongside the large cloud-computing
providers, but also frontier leaders such as Dario Amodei (CEO of Anthropic)
and Demis Hassabis, CEO of Google DeepMind.
c) Drivers
of a socially legitimized AI: they emphasize ethics, sustainability, and
social cohesion as conditions for legitimizing AI deployment. In Davos,
Germany, France, and Canada expressed themselves along these lines, emphasizing
society’s absorption of change. Friedrich Merz, Federal Chancellor of Germany,
for example, defended alliances “among equals” and a Europe antithetical to
unfair practices and commercial arbitrariness (Merz, 2026). For her part, the
President of the European Commission, Úrsula von der Leyen, not only noted that
the Union and India “are on the cusp of a historic trade agreement [that some]
call it the mother of all deals,”; she also made clear that Europe has all the
necessary capabilities to drive and attract investment in AI: “What we need is
to mobilise collectively these assets to their full potential. And to focus on
the essential. Focal point number one is to create a conducive and predictable
regulatory environment”.
Almost as
an anecdote, the President of France, Emmanuel Macron, wore blue aviator
glasses with reflective lenses throughout the Forum, even during his speech,
which seemed to reinforce the narrative of a Europe that not only legislates
technology but also inhabits it (although, in fact, they appear to have been
linked to an eye condition).
At the
corporate level, this orientation also appears in interventions that connect
technological legitimacy and sustainability, such as those of André Hoffmann,
Vice Chairman of Roche Holding and interim co-chair of the WEF, who warned that
“If you have no nature, you have no humanity, you have no business, you have no
dividend, you have no shareholders” (WEF, 2026h).
Conclusions
Davos 2026
showed that AI has an expanded political meaning: economic opportunity,
systemic risk, strategic infrastructure, and a vector of international
positioning; and it also showed that it is already a central object of informal
governance. While AI was integrated into a cross-cutting narrative and was
pointed to as an object of global competition and cooperation, the dispute over
its terms remains open. In Davos, divergences between Europe and the United
States on AI became visible in the context of the broader economic and
political tensions between both sides.
Davos 2026
helped delineate problems, acceptable risks, and plausible trajectories whose
advancement and systematization will be contested in formal international
arenas. In this sense, it operated as a device of cognitive alignment: it did
not determine outcomes per se, but it increased the visibility of certain
frames and vocabularies, facilitating their portability to other spaces.
Elite
positioning does not replace formal politics, but it can condition which
proposals appear reasonable, urgent, or inevitable, and the constellation
scheme we proposed captures that dynamic and the generation of cross-cutting
interests among different actors. There is no binary belonging (is part/is not
part) to a constellation; rather, there are recurrent patterns of argumentation
in clusters with undefined edges, which leads to agreement on the need to
establish an international regime for AI, being a consensus of form, but not of
content.
If these
constellations persist—and knowing that will require tracking positions in a
changing framework—global AI governance will tend less toward a “grand
agreement” and more toward partial, sectoral alignments. Future research should
track how the narratives we saw in Davos about AI, and the actors positioning
behind them, do or do not circulate into formal multilateral negotiations,
domestic regulatory agendas, and corporate standards, and assess how the
constellations respond to growing geopolitical rivalry.
And the
centrality of AI in Davos cannot be understood apart from today’s geo-economic
competition: its presence in the Global Risks Report 2026, alongside
geo-economic confrontation as the dominant risk, reinforces the perception of
AI as a strategic technology—an argument that can be used both to accelerate
and to slow its development (WEF, 2026a).
Finally,
this work suggests that spaces such as the WEF participate in the co-production
of the technological and international order, understood as a process of
coevolution among frames of legitimacy, action priorities, and criteria of
authority, built through interaction between public and private actors. That
process affects which problems become thinkable and legitimate and which actors
gain or lose authority, in a framework where innovation, power, risk, and
legitimacy are intertwined.
As a final
comment, we want to note that the human vision of technology had much less
prominence than the technological one, although it resonated in Julie Sweet’s
(2026), President and CEO of Accenture, phrase when she said, “It’s human in
the lead, not human in the loop”.
After all,
as Melvin Kranzberg (1986:557) said some time ago, “Technology is a very human
activity.” And discussing AI while losing its human focus is like being a
lawyer but scorning the law: possible, but contradictory and with potentially
serious effects.
References
Collingridge,
J., & Wearden, G. (2026, January 21). Jamie Dimon warns AI could move
faster than society can adapt. The Guardian. https://www.theguardian.com/technology/2026/jan/21/rollout-ai-slowed-save-society-jp-morgan-jamie-dimon-jensen-huang
Digital
Watch. (2026, January 23). Humanoid robots and AI take centre stage as Musk
joins Davos 2026. Geneva Internet Platform. https://dig.watch/updates/humanoid-robots-and-ai-take-centre-stage-as-musk-joins-davos-2026
Huang.
(2026 January 21). Conversation with Jensen Huang, President and CEO of
NVIDIA [Video]. World Economic Forum Annual Meeting 2026. https://www.youtube.com/watch?v=hoDYYCyxMuE
Jafar, B.
(2026, January 22). Remarks at World Economic Forum Annual Meeting 2026
(cit. en WAM y traducido por la IA). https://www.wam.ae/en/article/bycu8w0-global-leaders-call-for-responsible-deployment-fo
Kande, M.
(2026, January 19). PwC’s global chairman says most leaders have forgotten
‘the basics’ amid AI adoption challenges / Entrevistado por Diane Brady y
Nick Lichtenberg. Fortune. https://fortune.com/2026/01/19/pwc-global-chairman-mohamed-kande-ai-nothing-basics-29th-ceo-survey-davos-world-economic-forum/
Lifeng, H.
(2026, January 20). Davos 2026: Special Address by He Lifeng, Vice-Premier of
the People's Republic of China, to the World Economic Forum [Video]. World
Economic Forum. https://www.youtube.com/watch?v=lqnsjRnizOE&t=287s
Merz, F.
(2026, January 22) Davos 2026: Special Address by Friedrich Merz, Federal
Chancellor of Germany to the World Economic Forum [Video], World Economic
Forum. https://www.youtube.com/watch?v=UVwPUnAhis0&t=2s
Parmelin,
G. (2026, January 13). Davos 2026: Special address by Guy Parmelin, President
of the Swiss Confederation [Transcript]. World Economic Forum. https://www.weforum.org/stories/2026/01/davos-2026-welcoming-remarks/
Subianto,
P. (2026, January 22). Special Address by Prabowo Subianto, President of
Indonesia [Transcript/Video]. World Economic Forum Annual Meeting 2026. https://www.weforum.org/meetings/world-economic-forum-annual-meeting-2026/sessions/special-address-by-prabowo-subianto-president-of-indonesia/
Sweet, J.
(2026, January 20). Participation at the 2026 WEF event Scaling AI: Now Comes
the Hard Part [Video]. World Economic Forum Annual Meeting 2026. https://www.weforum.org/meetings/world-economic-forum-annual-meeting-2026/sessions/scaling-ai-now-comes-the-hard-part/
Trump, D,
(2026, January 21) Davos 2026: Special Address by Donald J Trump, President of
the United States of America [Transcript]. World Economic Forum. https://www.weforum.org/stories/2026/01/davos-2026-special-address-donald-trump-president-united-states-america/
von der
Leyen, U. (2026, January 19). Special Address by the President von der Leyen:
World Economic Forum. European Commission. https://ec.europa.eu/commission/presscorner/detail/en/speech_26_150
Wearden, G.,
& Stewart, H. (2026, January 23). Young will suffer most when AI ‘tsunami’
hits jobs, says head of IMF. The Guardian.
https://www.theguardian.com/technology/2026/jan/23/ai-tsunami-labour-market-youth-employment-says-head-of-imf-davos
World
Economic Forum. (2026). World Economic Forum annual meeting 2026. https://www.weforum.org/meetings/world-economic-forum-annual-meeting-2026/
World
Economic Forum. (2026a). Global Risks Report 2026. https://www.weforum.org/publications/global-risks-report-2026/
World
Economic Forum. (2026b). Chief Economist's Outlook: January 2026. https://www.weforum.org/publications/chief-economists-outlook-january-2026/
World
Economic Forum. (2026c). Four futures for jobs in the new economy: AI and
talent in 2030. https://reports.weforum.org/docs/WEF_Four_Futures_for_Jobs_in_the_New_Economy_AI_and_Talent_in_2030_2025.pdf
World
Economic Forum. (2026d). New economy skills: Building AI, data and digital
capabilities for growth. https://www.weforum.org/publications/new-economy-skills-building-ai-data-and-digital-capabilities-for-growth/
World
Economic Forum. (2026e). Proof over Promise: Insights on Real-World AI
Adoption from 2025 MINDS Organizations. https://www.weforum.org/publications/proof-over-promise-insights-on-real-world-ai-adoption-from-2025-minds-organizations/
World
Economic Forum. (2026f, January 18). What is planetary intelligence and how
could it move AI from the internet to the real world? Weforum https://www.weforum.org/stories/2026/01/planetary-intelligence-ai-internet-real-world
World
Economic Forum. (2026g, January 20). Conversation with Satya Nadella, CEO of
Microsoft [Video]. World Economic Forum Annual Meeting 2026. https://www.weforum.org/meetings/world-economic-forum-annual-meeting-2026/sessions/conversation-with-satya-nadella-ceo-of-microsoft/
World
Economic Forum. (2026h, January 20). Building a business case for nature. Weforum. https://www.weforum.org/stories/2026/01/live-from-davos-2026-what-to-know-on-day-2/
This is the
original version of the article.
Spanish
version (ES) available here.