In the ever-evolving landscape of artificial intelligence, an intriguing development has emerged: the marriage of philosophy and AI. This unexpected alliance is not just a theoretical exercise but a practical one, with major tech companies and universities alike recognizing the value of philosophical insights in shaping the future of AI.
The Rise of Philosophical AI
Philosophy, often seen as an impractical pursuit, has found a new and powerful application in the world of AI. Tech investors like Paul Graham, who initially studied philosophy, eventually switched to AI, perhaps unknowingly setting a precedent for the industry's future. Today, philosophers with PhDs are being recruited by major tech companies to help navigate the ethical dilemmas posed by AI.
The numbers speak for themselves: in 2013, just 1% of roles on PhilJobs, the field's primary job board, were AI-related. Fast forward to last year, and that figure skyrocketed to 16%. This shift reflects a broader trend: the recognition that AI, while powerful, requires ethical guidance, and who better to provide that than philosophers, experts in moral dilemmas?
The AI-Philosophy Nexus
The relationship between AI and philosophy is not a one-way street. While philosophers are helping AI companies build more virtuous machines, AI is also influencing philosophical thought. Nick Bostrom's book, Superintelligence, for instance, has not only brought attention to the potential dangers of all-powerful AI but has also influenced research agendas across major labs.
Companies like OpenAI and Anthropic are taking this seriously. OpenAI consulted hundreds of moral philosophers and tech-ethics experts when designing rules for ChatGPT's behavior. Anthropic, meanwhile, has published an 84-page philosophical treatise outlining its intentions for its bot, Claude. This document, a true philosophical treatise, is then used to train Claude, ensuring its 'good character'.
Applied Philosophy
The work of philosophers in AI companies is often described as 'applied philosophy'. This term, coined by Amanda Askell, a philosopher at Anthropic, reflects the practical nature of their work. Philosophers are not just theorizing about AI; they are actively involved in its development, training AI systems to reason more rigorously about ethics.
This applied philosophy is not limited to full-time employees. AI companies are also hiring philosophy PhDs as consultants, offering up to $60 an hour for their expertise. These consultants feed ethical dilemmas into AI systems and evaluate the logic of their responses, a true fusion of philosophy and technology.
The Future of AI and Philosophy
The alliance between AI and philosophy is not without its challenges. Some philosophers worry about the distorting effect of the AI fervor on their discipline, fearing a rush of low-quality work driven by misaligned incentives. Others question the frenetic pace of AI development, which contrasts sharply with the slow, careful thought that philosophy encourages.
Despite these challenges, the future looks bright for this unusual partnership. New academic programs are springing up to train students in applied philosophy skills, and the academic job market is rewarding AI-related philosophy. The question remains: can AI truly do philosophy better than humans? This, as David Chalmers, a leading philosopher of mind, puts it, is a very interesting philosophical question indeed.
