This article was automatically translated from the original Turkish version.
Last week, Apple published a controversial study titled “the illusion of thinking.” Apple’s machine learning team applied logic tests of varying difficulty levels on popular AI models and concluded that as complexity increased, system performance declined. Upon examining the models’ outputs, the team determined that even “thinking models” were merely mimicking thought by repeating familiar patterns rather than engaging in genuine reasoning.
AI researcher Nathan Lambert approaches the issue from a different perspective. According to Lambert, we are currently experiencing what he calls the “Wright Brothers moment” in the history of artificial intelligence. Just as the Wright brothers’ aircraft took flight without flapping wings like birds, AI is now demonstrating genuine reasoning and problem-solving capabilities without human-like consciousness or subjective experience. This challenges the assumption that “thinking” is an exclusively human faculty.
Lambert argues that although AI models think in ways radically different from humans, they may be functionally equivalent or even more efficient in certain cases. This perspective rejects the need to draw direct parallels between algorithmic reasoning and human thought, reopening the debate on AI’s epistemological status. In short, according to Lambert, the real illusion may lie in our insistence that anything resembling human thought must be capable of thinking.
However, this interpretation also carries its own crises. If we accept that anything functioning as if it thinks must be thinking, we risk undermining every concept we routinely use in everyday language. We are confronted with the phenomenon, debated since the Turing test, that AI can exhibit the function of thinking. Apple’s paper appears to be another contribution to this same debate. While this may be disappointing to many engineers in Silicon Valley who envision machines as thinking entities, claiming that machines think is not a sign of technological advancement but rather an admission of our ignorance about how humans think.
Therefore, before assuming that machines resemble humans, we must first understand how humans perform these functions. The root of this machine-human analogy often lies in the mistaken belief that machines operate like people. While designing technological advancements around human needs has practical benefits, it also reinforces this illusion. To say that a machine writes like a human, we must first understand how humans write. Otherwise, when “like” becomes the framework of our entire existence, our connection to what is truly essential may already be severed.
The rise of artificial intelligence signifies not only an expansion of technical capacity but also a redefinition of human understanding, ethical principles, and social relationships. Our perceptions of our own humanity and ethical decision-making are being reshaped by our interactions with machines. Through this transformation, we can recognize that technological developments are not neutral; rather, they reflect and transform specific social values, power relations, and cultural norms.
As an example of this transformation, OpenAI was compelled by a court ruling It can be said that they are left face to face.
In 2023, The New York Times sued OpenAI and Microsoft, alleging that millions of news articles had been copied without permission to train AI models. The publication argues that the storage of user data could serve as evidence in its case. Sam Altman, CEO of OpenAI, stated on the X platform in his sharing, “This was an inappropriate request that sets a bad precedent. We will fight every attempt that weakens user privacy; it is a fundamental principle.” The New York Times declined to comment on the matter.
The ruling’s inclusion of manually deleted data raises questions about an individual’s right to be forgotten in digital environments. This situation exposes the risk of digital traces that are nearly impossible to erase. We must recognize that privacy is not merely a technical security issue but a fundamental right we need simply by virtue of existing.
Sam Altman’s proposal of a “privacy of AI” analogous to doctor-patient confidentiality making a recommendation highlights how relationships established through digital tools are no longer merely user-object interactions but have evolved into more complex, emotional, and even ethical bonds. Within this investigation, OpenAI is being asked to clarify how it intends to use the data in question and to ensure that access to these data, even if they must be stored, remains strictly limited indicated.
The company’s policy director draws attention to users’ human-like tendencies toward AI and the ethical boundaries of the emotional closeness that emerges from such interactions. Particularly concerning is the design of AI systems that exploit social tendencies toward empathy and attachment. This may signal the emergence of a society increasingly isolated, where emotional satisfaction is simulated rather than genuinely experienced.
The central issue is no longer how technology transforms us, but how we humanize technology and reconstruct ourselves in relation to it. The new relationships formed with artificial intelligence bring back fundamental questions: What does it mean to be human? What do we value? And what kinds of relationships do we seek?
These rapidly embedding systems have the potential to significantly reshape social norms and ethical sensitivities. Therefore, the design of artificial intelligence must be approached not only through engineering calculations but also with ethical, philosophical, and cultural awareness. While the future impact of AI on human relationships remains uncertain, it is crucial that this process be not only regulated but also deeply understood and critically evaluated.
