Cormody Gray, Professor of Catholic Theology at Durham University in the UK and a prominent figure in the fields of philosophy of science and ethics, recently published a note explaining his reasons for rejecting a collaboration offer from the company Anthropic.
According to ETNA citing the Financial Times, Gray explains in this article that the AI industry increasingly uses human terms such as “intelligent” and biological and neuroscientific concepts to describe large language models; a trend that, he says, is even seen in research reports of companies like Anthropic and OpenAI, including Anthropic’s research paper titled “On the Biology of a Large Language Model.”
Why the offer was rejected
Based on reports published about this article, Gray decided to refuse collaboration with Anthropic because, in his view, the AI industry often uses the humanities merely as a decorative tool and to improve its public image, not for real accountability.
He warns that humanities researchers, by working closely with tech companies, are at risk of losing their academic independence and would be better off focusing their attention on human agency rather than getting involved in abstract debates about the nature of artificial intelligence.
The main philosophical argument
In the central part of his article, Gray argues that describing large language models with biological terms is a category mistake.
He believes that genuine consciousness and intelligence are not merely information processing, but are rooted in a biological existence that experiences its needs, fears, and desires over time and place. From this perspective, the similarity between the human brain and computational systems is far outweighed by their differences, and for this reason, using the label “artificial intelligence” can be regarded as a misleading and consequential misnaming that distorts the true nature of this technology.
Importance of the issue
This note was published at a time when tech giants including Anthropic and OpenAI are increasingly turning to humanities and philosophy experts to bolster the social and ethical legitimacy of their products.
Gray’s criticism reflects the concern of a part of the academic community that collaborating with large tech companies poses a threat to researchers’ intellectual independence.