Large language models that power AI should be publicly owned | Letter

TruthLens AI Suggested Headline:

"Call for Public Ownership of Large Language Models in Historical Research"

Large language models that power AI should be publicly owned | Letter analysis cover image The Guardian 7.9
Image Source: https://i.guim.co.uk/img/media/12dee83cbee13f8b3c19192aa9ffc14879f5b3c6/130_0_4569_3655/master/4569.jpg?width=1200&height=630&quality=85&auto=format&fit=crop&overlay-align=bottom%2Cleft&overlay-width=100p&overlay-base64=L2ltZy9zdGF0aWMvb3ZlcmxheXMvdGctZGVmYXVsdC5wbmc&enable=upscale&s=b7230d600b03859574cd71974f3b5c9c
Original Article Source: https://www.theguardian.com/technology/2025/may/26/large-language-models-that-power-ai-should-be-publicly-owned
View Raw Article Source (External Link)
Raw Article Publish Date:
AI Analysis Average Score: 7.9
These scores (0-10 scale) are generated by Truthlens AI's analysis, assessing the article's objectivity, accuracy, and transparency. Higher scores indicate better alignment with journalistic standards. Hover over chart points for metric details.

TruthLens AI Summary

Large language models (LLMs) have recently revolutionized the field of historical research by enhancing the ways in which scholars process, annotate, and generate texts. Despite their transformative potential, historians are prompted to critically evaluate the ownership of these powerful tools and the implications they have for our understanding of history. Currently, most dominant LLMs are developed by private corporations, whose primary motivations often revolve around profit and intellectual property control. This corporate focus can conflict with the fundamental values of historical scholarship, which emphasize transparency, reproducibility, accessibility, and cultural diversity. The reliance on these private entities creates significant concerns regarding the opacity of the models, as researchers frequently lack clarity on the training data and inherent biases embedded within them. Additionally, the instability of access and the potential for abrupt changes in terms of service can hinder research efforts, particularly for those in less-resourced environments, leading to inequity in the availability of these advanced tools.

To address these challenges, there is a pressing need to advocate for the development of public, open-access LLMs specifically tailored for the humanities. Such models should be built using curated, multilingual, and historically grounded corpuses sourced from libraries, museums, and archives, ensuring that they are transparent and accountable to the academic community. The creation of these public resources, while undoubtedly complex, is essential for fostering an environment where scholarly integrity can thrive. Just as national archives and educational curricula are not outsourced to private entities, the tools that shape our interpretive technologies should also remain in the public domain. The humanities hold both a responsibility and a unique opportunity to cultivate culturally aware and academically grounded artificial intelligence. By advocating for public ownership of LLMs, scholars can ensure that the future of public knowledge remains robust and accessible to all.

TruthLens AI Analysis

The article presents an argument for the public ownership of large language models (LLMs) used in artificial intelligence, particularly concerning their application in historical research. It raises critical questions about the ownership and control of these powerful tools, emphasizing the need for transparency, equity, and academic integrity in their development and use.

Ownership and Control of LLMs

The author, Prof. Dr. Matteo Valleriani, highlights that the most impactful LLMs are currently developed by private companies, driven by profit motives rather than the principles of scholarship. This situation creates a tension between the goals of historical research and the operational priorities of these corporations, which often do not align with the values of the academic community.

Concerns Raised

The article outlines several concerns, including opacity regarding training data and biases, instability of access, and inequity faced by researchers in less-resourced environments. These points underscore the potential risks of relying on privately owned technologies for scholarly work.

Call for Public, Open-Access Models

Valleriani advocates for the establishment of public, open-access LLMs tailored for the humanities. He argues that such models should be based on curated, multilingual data from cultural institutions and should prioritize transparency and accountability to the academic community.

Cultural Responsibility and AI

By framing the development of LLMs as a cultural responsibility, the article emphasizes the potential for creating AI that is not only technically competent but also culturally aware. This perspective suggests a proactive approach to AI that aligns with the ethical standards of historical scholarship.

Perception and Strategy

The article likely aims to foster a sense of urgency among historians and scholars about the implications of private control over essential research tools. It encourages public discourse on the issue, motivating stakeholders to advocate for public ownership and funding in AI development.

In terms of manipulation, the article seeks to frame the narrative around LLMs in a way that promotes public ownership as a moral imperative. While it does not appear to contain overt manipulation, the language used underscores the importance of equity and transparency, which could evoke strong responses from readers who share these values.

The reliability of the article stems from its alignment with ongoing discussions in the academic community about the implications of AI on scholarship. By citing specific concerns and outlining a clear vision for the future, the article contributes meaningfully to the conversation about the role of technology in humanities research.

By emphasizing responsibility and accountability in AI development, it potentially resonates with communities invested in cultural heritage and educational integrity. This could attract support from academic institutions, cultural organizations, and advocacy groups focused on public access to knowledge.

As for the economic and political implications, advocating for public ownership of LLMs could influence funding policies and shape the future landscape of AI development. It may provoke discussions about regulation and the role of private companies in public scholarship.

In summary, the article presents a compelling case for the need to rethink the ownership and development of LLMs within the humanities. While it raises valid concerns and advocates for a more equitable approach, it does so in a manner that encourages public engagement and critical discourse on the future of AI in academic settings.

Unanalyzed Article Content

Source URL: https://www.theguardian.com/technology/2025/may/26/large-language-models-that-power-ai-should-be-publicly-owned

Large language models (LLMs) have rapidly entered the landscape of historical research. Their capacity to process, annotate and generate texts is transforming scholarly workflows. Yet historians are uniquely positioned to ask a deeper question – who owns the tools that shape our understanding of the past?

Most powerful LLMs today are developed by private companies. While their investments are significant, their goals – focused on profit, platform growth or intellectual property control – rarely align with the values of historical scholarship: transparency, reproducibility, accessibility and cultural diversity.

This raises serious concerns on a) opacity: we often lack insight into training data and embedded biases, b) instability: access terms and capabilities may change without notice, and c) inequity: many researchers, especially in less-resourced contexts, are excluded.

It is time to build public, open-access LLMs for the humanities – trained on curated, multilingual, historically grounded corpuses from our libraries, museums and archives. These models must be transparent, accountable to academic communities and supported by public funding. Building such infrastructure is challenging but crucial. Just as we would not outsource national archives or school curriculums to private firms, we should not entrust them with our most powerful interpretive technologies.

The humanities have a responsibility – and an opportunity – to create culturally aware, academically grounded artificial intelligence. Let us not only use LLMs responsibly but also own them responsibly. Scholarly integrity and the future of public knowledge may depend on it.Prof Dr Matteo VallerianiMax Planck Institute for theHistoryof Science, Berlin, Germany

Have an opinion on anything you’ve read in the Guardian today? Pleaseemailus your letter and it will be considered for publication in ourletterssection.

Back to Home
Source: The Guardian