Tech • IA • Crypto
10 AI founders: their bets for 2026 and their fears!
7/10TL;DR
European tech leaders warn that data sovereignty, enterprise AI control, and shifting cloud economics are becoming decisive factors in global competitiveness.
KEY POINTS
Scality’s global positioning
French data storage firm Scality, founded 15 years ago with roots in Paris, San Francisco, and Tokyo, has grown into a global leader serving critical infrastructure. The company manages data for governments in 35 countries, major hospitals, and 15 of the world’s largest banks, where system failures could halt surgeries or financial operations. Its technology also underpins Iron Mountain’s Iron Cloud in the United States.
European sovereignty gap
Despite strong capabilities, European firms struggle to secure domestic public contracts. Unlike the United States, which enforces mechanisms such as the Small Business Act, or countries like India and Saudi Arabia that favor local production, Europe lacks coordinated industrial policy. Public buyers often assume contracts awarded to large integrators like Capgemini or Atos support local ecosystems, even when underlying technologies are sourced from IBM or Dell.
Cost and cloud repatriation trends
Scality claims its software-defined storage solutions are about 30% cheaper than negotiated hyperscaler pricing. This cost gap is driving a shift among large enterprises away from fully outsourced cloud models. While small and mid-sized firms remain dependent on hyperscalers, large organizations are increasingly exploring hybrid or repatriated infrastructure strategies to regain control and reduce expenses.
AI as a systemic transformation
Artificial intelligence is described as a disruption on par with the internet and mobile computing, with deep implications for social behavior and economic structures. Beyond productivity gains, AI is expected to redefine how individuals interact, make decisions, and organize work. The long-term societal effects remain uncertain, echoing earlier underestimations of social media’s impact.
Enterprise control of AI systems
Companies are increasingly concerned about losing control over internal knowledge when using external AI platforms. Enterprise AI providers such as Glean emphasize the importance of retaining ownership of “organizational memory,” including workflows, expertise, and decision-making patterns. This data is seen as a strategic asset that must remain under company control rather than embedded within third-party models.
Rise of specialized AI agents
A shift is underway from general-purpose AI tools to task-specific agents embedded in business processes. These agents can automate functions such as contract review, customer support, and sales workflows, significantly improving productivity. Their effectiveness depends less on the underlying model and more on access to proprietary company data and contextual knowledge.
Commoditization of AI models
As foundational models become more accessible, differentiation is moving away from model performance toward data, orchestration, and integration layers. Companies that control unique datasets and workflows are better positioned to maintain competitive advantage, while reliance on identical external models risks commoditizing outputs across firms.
Geopolitical fragmentation of AI
Access to advanced AI models is increasingly shaped by geopolitical constraints. Some cutting-edge systems are restricted to specific nationalities or regions, reinforcing concerns about technological dependency. Meanwhile, Chinese open-weight models are gaining traction globally, challenging Western dominance and complicating efforts to maintain technological leadership.
New economic logic in software
The emergence of AI-driven systems is reshaping valuation models in the software industry. Traditional metrics such as revenue multiples are giving way to long-term survivability and adaptability. Companies like Datadog, Snowflake, and Figma are cited as examples of firms repositioning themselves as infrastructure layers within AI ecosystems, rather than standalone applications.
On-premise infrastructure revival
A partial return to on-premise or dedicated infrastructure is emerging, particularly for AI workloads requiring GPU-intensive computing and large energy capacity. Instead of traditional in-house data centers, companies are expected to lease dedicated, private environments combining hardware, power, and storage while maintaining operational control.
CONCLUSION
Control over data, infrastructure, and enterprise-specific AI capabilities is becoming central to economic power, with Europe facing structural challenges in asserting technological sovereignty amid intensifying global competition.