More Things in Heaven and Earth

There is an old joke about the philosophy degree as a fast track to unemployment. AI is changing that. One philosophy graduate inverted the joke , saying this is “probably the best time to be a philosopher since Aristotle.” Anthropic’s Amanda Askell, a philosopher by training, is one of the company’s most recognizable figures. Google DeepMind employs an in-house ethicist. Why the sudden rush to frontier lab bull sessions? The answer is found in the vague, overused concept called “ alignment .” In plain English, it is the increasingly frantic effort to make sure that LLMs and their agent progeny do not spin out of control and drown us in paperclips . Alignment seeks to give frontier AI systems values, a slippery task that usually raises the corrosive, relativistic question, “says who?” which tends to derail most modern conversations about universal ethics. Applied morality and ethics are not engineering issues. They are the questions philosophy departments have worked on for 25 centuries. All that was old is startlingly new again. From a practical standpoint, the rise of the philosophers is the paradigmatic example of the new judgment economy. As AI absorbs routine cognitive tasks, workflow bottlenecks shift from task execution to design and oversight. David Deming’s 2017 study in the Quarterly Journal of Economics found that occupations requiring both cognitive and social skills saw the strongest employment and wage growth between 1980 and 2012. The philosopher at an AI lab is the leading edge of the same pattern. The more capable the machine, the more valuable the person who can reason about ends and means simultaneously. Before the class of 2030 rushes to declare philosophy majors, a caution is in order: frontier labs will not absorb thousands of new philosophy degrees. What students and workers ought to be keeping an eye on is how philosophical questions and knowledge are likely to occupy more of their time and brain power in an AI-infused economy and invest accordingly. The “can we do it, should we do it” dilemma takes on more urgency when powerful new tools increase human capacity for good and evil. For employers and their customers alike, ethical definitions and applications are also bottom-line decisions. A few high-profile mistakes (cf. self-driving car fatalities) have public trust and profit implications. Ergo, companies building and using advanced technology are likely to place growing value on a skill set the labor market spent two decades telling everyone to avoid. The lesson for students, parents, and policymakers who shape education funding is not that everyone should major in philosophy. It is that the capacity for judgment, formed through liberal education that teaches people how to grapple effectively with hard questions, is becoming a market asset rather than a luxury good. Nick Bostrom. Mr. Paperclip himself, warned in 2014 that machine intelligence gives us “philosophy with a deadline.” The deadline is now working its way into job descriptions. The post More Things in Heaven and Earth appeared first on American Enterprise Institute - AEI .