Understand the Risks of Big Tech Dependency in Modern Newsrooms
Reading news online without the intermediation of US technology firms has become practically impossible. Audiences email through corporate systems, search with dominant engines, and scroll through algorithmic social feeds. Dutch news media rely heavily on these same companies to distribute content, host data, and operate daily. Hannes Cools, an Assistant Professor of Communication Science at the University of Amsterdam, notes that news organizations are effectively hooked up to a digital drip. When that drip remains in foreign hands, it creates structural vulnerabilities that affect both the media sector and the broader democratic landscape.
Addressing this issue requires a critical examination of how news infrastructure operates today. The digital autonomy of journalism is no longer a niche technical concern; it is a foundational requirement for editorial independence. As highlighted in recent UvA News Articles, researchers and progressive news organizations are actively building alternative digital infrastructures to reduce this reliance. Understanding the exact nature of these risks is the first step toward mitigating them.
Analyze the Economic and Democratic Threats to Netherlands Journalism
Dependence on US Big Tech companies introduces two primary categories of risk: democratic and economic. On the democratic front, external commercial entities such as Google, Microsoft, Meta, and OpenAI hold significant power over information access. A striking example of this vulnerability occurred at the International Court of Justice in The Hague, where a judge suddenly lost access to his email account because Microsoft blocked him. While this specific incident happened within the judiciary, the underlying mechanism applies equally to journalism. If a tech company can unilaterally restrict access to communication tools, they possess the ability to hinder investigative reporting or silence specific editorial voices.
Economically, the current ecosystem actively drains revenue and traffic from traditional news outlets. Search engines and social platforms increasingly build proprietary AI layers on top of search results. While these features offer convenience for users, they directly reduce the number of people who click through to the original news websites. Social media companies contribute to this decline by prioritizing user retention within their own closed ecosystems rather than referring audiences outward. Analysts expect visitor numbers to news sites to drop sharply in the coming years as these AI integrations expand.
Protecting the watchdog role of the press in the Netherlands means confronting these economic realities. Without direct access to their audiences, newsrooms cannot secure the subscriptions or advertising revenue required to fund quality journalism.
Identify the Structural Weaknesses in Current Media Infrastructure
Many news organizations lack sufficient control over their own data storage, content distribution, and audience analytics because these functions are entirely outsourced to American tech conglomerates. This outsourcing creates a single point of failure. A change in a platform’s terms of service, an algorithmic tweak, or an account suspension can immediately disrupt a newsroom’s operations. Building structural resilience requires mapping these dependencies and identifying viable alternatives that align with journalistic values rather than corporate profit motives.
Explore European and Open-Source Alternatives for News Media
Achieving digital autonomy does not mean news organizations must build every technical solution entirely in-house. Developing proprietary infrastructure from scratch would be prohibitively expensive and inefficient for most newsrooms. Instead, the focus shifts to identifying and adopting European and open-source providers that better align with the ethical standards of the press.
A central component of this transition involves the use of open-source Large Language Models (LLMs). An LLM is a computer program utilizing artificial intelligence to understand, predict, and generate human language. Unlike commercial products like ChatGPT, an open-source LLM makes its underlying code publicly available. This transparency offers several distinct advantages for news organizations:
- Security and Verification: Technical teams can independently verify that the software is secure and free from hidden vulnerabilities.
- Privacy Protection: Organizations can confirm exactly how the model handles data, ensuring it does not harvest personal user information.
- Editorial Alignment: Developers can adapt and fine-tune the model to adhere strictly to journalistic codes of conduct, preventing the dissemination of hallucinated or biased information.
By leveraging these open-source tools, newsrooms can implement modern AI features without surrendering their editorial integrity or audience data to third-party corporations.
Share your experiences with AI and news consumption in the comments below.
Review the UvA Research Approach to Digital Autonomy
The University of Amsterdam is actively facilitating this transition through a structured, two-year research project initiated in 2026. Led by Hannes Cools from the AI, Media & Democracy Lab, the project brings together a multidisciplinary consortium designed to tackle the technical, legal, and practical challenges of digital independence.
Collaborate Across Disciplines and Organizations
The project relies on a robust partnership network. The UvA Faculty of Law contributes critical expertise regarding regulation, digital infrastructure, and licensing. The Amsterdam University of Applied Sciences (AUAS) provides deep technical knowledge in developing and deploying open-source language models. Crucially, the project includes direct practice partners: the think tank Waag, and the news platforms De Correspondent and Follow the Money (FTM). These organizations possess a demonstrated need for digital autonomy. By directing grant funding to both the university and these newsrooms, the project ensures high motivation for continuous, practical collaboration.
Implement a Three-Phase Research Design
The research team structured the initiative into three distinct phases to ensure actionable outcomes:
- Definition and Mapping: Researchers work with practice partners to clearly define what digital autonomy means for their specific operations. They map current dependencies on Big Tech and catalog available European and open-source alternatives.
- Public Engagement: The team conducts focus groups with diverse participants, varying by age, AI experience, and readership habits. These sessions gather public input on the definition of digital autonomy and establish expectations for how AI should function within a news context.
- Prototype Development: In the final phase, the news organizations build functional prototypes for readers to test. Users interact with the newly selected or purpose-built AI models to perform tasks such as searching article archives, generating summaries, or listening to audio versions of stories.
Explore our related UvA News Articles for further reading on media innovation.
Assess the Future of Independent Journalism Infrastructure
Readers will likely notice the impact of this shift through increased fairness and transparency in how news organizations utilize artificial intelligence. Rather than opaque, black-box algorithms dictating what audiences see, open-source implementations allow for clear explanations of how content is filtered, summarized, or presented. Furthermore, practical features like archive search, article summarization, and text-to-speech functionality will become more reliable and privacy-friendly.
The long-term success of this initiative depends on broader industry adoption. While De Correspondent and Follow the Money serve as vital testing grounds for these concepts, true digital autonomy requires larger media conglomerates to follow suit. If major publishers in the Netherlands and across Europe begin transitioning to sovereign digital infrastructures, the journalism industry as a whole can reclaim control over its work. This structural shift is necessary to preserve the press’s role as an independent democratic watchdog, free from the arbitrary influence of foreign tech monopolies.
Have questions about implementing open-source models in your organization? Write to us!
