By Stella Kyratzi, Postdoctoral Researcher at KIN center for Digital Innovation, Vrije Universiteit Amsterdam, Phuong Hoan Le, Assistant Professor Organization & Management, Erasmus University Rotterdam, and Pierrette Howayeck, Postdoctoral Researcher at ESSCA School of Management
This blog is produced as a part of the Reshaping Work Fellowship Programme. The opinions and views expressed in this publication are those of the author. They do not purport to reflect the opinions or views of Reshaping Work or organisations that have supported the programme.
This month, our Reshaping Work Fellows book club discussed Empire of AI: Dreams and Nightmares in Sam Altman’s OpenAI (2025) by Karen Hao. Given the title, one might expect a history of OpenAI, the organization that launched ChatGPT, and the book indeed offers a wealth of insider knowledge. Yet it would be simplistic to see it merely as a corporate account of Silicon Valley drama. As Hao cautions her readers, “While [this book] tells the inside story of OpenAI, that story is meant to be a prism through which to see far beyond this one company.” Using ‘empire’ as a metaphor to hint at AI colonialism, the book raises urgent questions about the global tech arms race and the ethical, social, and environmental costs of a logic that prizes unprecedented scale and consumption. In this post, Stella Kyratzi, Pierrette Howayeck, and Phuong Hoan Le touch upon three interrelated themes that animated our discussion: how AI redefines intelligence, how Big Tech rewrites trust and ownership, and how corporate actors reframe the very idea of progress.
It’s About Who Decides What “Intelligence” Means
– Stella Kyratzi
“In a vacuum of agreed upon meaning, Artificial Intelligence or Artificial General Intelligence can be whatever OpenAI wants.”
This line from Empire of AI cuts to the heart of today’s technological frenzy. In the absence of a stable definition of intelligence, the term becomes elastic — moulded to fit corporate ambition, investor optimism, or ideological faith. OpenAI, as Karen Hao reveals, is not just building technology; it is building meaning. Intelligence, in this empire, is whatever can be computed, scaled, or monetized. The irony is that while AI promises to replicate or even surpass human intelligence, we still do not know what intelligence itself truly is. We are thus trying to outdo something we cannot fully define — a strange act of technological hubris cloaked as progress?
As philosopher Hubert Dreyfus warned decades ago, intelligence is not a set of abstract rules or data patterns to be imitated — it is an embodied, situated way of being in the world. Humans do not just process information; we interpret, feel, hesitate, and care. Machines, by contrast, are brilliant pattern recognizers without understanding, executing correlations without consciousness. Yet, the empire of AI turns this lack of understanding into a strength: by redefining intelligence as prediction, it reclaims ignorance as efficiency.
Perhaps, then, the real question is not how close AI is to intelligence, but who gets to define intelligence at all. In today’s AI economy, power over that definition shapes the direction of research, the narratives of inevitability, the futures we imagine possible, and the sacrifices we are willing to make along the way. To ask “who knows what AI really is” is to pierce the illusion that intelligence is a neutral object waiting to be replicated. It is, instead, a contested idea: one that reveals as much about our social desires and fears as it does about machines themselves.
The GitHub Heist: How Big Tech Broke the Internet’s Gift Economy
– Pierrette Howayeck
“While giving OpenAI free access to the code in GitHub’s public repositories was not illegal, it still felt like a violation of the user community’s trust. Much of that code had been shared in the spirit of fostering open-source software development, which was grounded in helping independent developers and small startups have a chance at being competitive, not in helping the big players entrench their monopoly.”
This passage from the Empire of AI reveals a pivotal moment when corporate ambition collided with community trust. GitHub grew into a space where developers shared their work freely, believing in building something bigger together. Anthropologists like Marcel Mauss call this a “gift economy”, where people give without expecting immediate returns because they believe that the system will benefit everyone.
In 2018 Microsoft acquired GitHub and in 2020 it allowed OpenAI to take all the shared work to train its models. Some of Microsoft’s employees felt uncomfortable, recognizing it as a betrayal of community trust. Yet, such training was argued to be “an important stepping stone” towards OpenAI’s definition of Artificial General Intelligence — “what better way to achieve OpenAI’s mission?” Classic ends-justify-the-means thinking. If you’re saving humanity, apparently you can take whatever you need.
The irony here is painful. Researchers used decades of collaborative work to build systems that now may replace the very people who created that work. We already see this with so-called vibe coding; where AI tools generate functional code from simple prompts, making programming accessible to non-coders. It’s like using someone’s recipe collection to open a restaurant that puts them out of business. Yet, people didn’t stop sharing, and even developers who understand how these AI tools were trained use them because these have become integral to modern workflows.
The GitHub case shows how AI development might significantly change knowledge sharing. Alternatives exist: the EU AI Act mandates transparency and accountability, while UNESCO’s AI Commons proposes frameworks where communities retain control over their contributions. We can either build AI that fosters human collaboration while ensuring that communities benefit from their contributions, or allow for companies to consume the relationships that make innovation possible. This choice will shape both AI’s future and how humans work together.
Reclaiming the Future: Challenging Corporate Framing of Progress
– Phuong Hoan Le
As a scholar in corporate communication and management, what stuck with me most from Hao’s Empire of AI is the power of framing. The book is filled with examples of how corporations shape media narratives, influence policy debates, and steer public opinion through PR campaigns and targeted lobbying. Even the name “artificial intelligence” was originally a marketing tool for scholars: it was chosen because the initial label of “automata studies” did not attract enough attention. As Stella Kyratzi already touched upon, OpenAI and its competitors wield large power over public understanding, framing Artificial General Intelligence (AGI) as the inevitable path forward. Not only does the term “AGI” lend itself well to exaggerations about what these technological innovations may be able to do one day, anthropomorphism also helps corporations reframe the stealing of data from artists and authors as merely ‘inspiration,’ just like humans are inspired by each other’s work.
Within the context of AI colonialism, Hao explains how OpenAI’s founding mythology of building AGI that “benefits all of humanity” is used to justify their race to build the technology quickly, before another entity builds a bad version of it. Ethical, social, and environmental considerations are deprioritized, because the hypothetical AGI can solve problems such as climate change. Seen in this light, the GitHub case is not only about code but also narrative power: once exploitation can be framed as progress, the practice becomes easier to normalize, scale, and defend.
Taken together, the book and our discussion suggest that AI is not merely a technical field but a political-institutional project: one that defines what counts as intelligence, appropriates the commons that sustained it, and storyboards the future. If we want a more equitable and sustainable future for tech — one that is not rooted in extraction but in collective responsibility and care — we will have to challenge those frames and wrest back control of the story about what progress should look like. As Hao wrote:
“Artificial intelligence doesn’t have to be what it is today. We don’t need to accept the logic of unprecedented scale and consumption to achieve advancement and progress. […] But the empires of AI won’t give up their power easily. The rest of us will need to wrest back control of this technology’s future. And we’re at a pivotal moment when that’s still possible.