Mathematician and scientist Stephen Wolfram grew up in a household where his mother was a professor of philosophy at Oxford University. This led him to not want to get involved in philosophy when he was younger, but as he got older and wiser, Wolfram saw the value in thinking things through deeply. Now he wants to bring some of that deep philosophical rigor to his AI research, to help us better understand the problems we face as AI becomes more capable.
Wolfram was something of a child prodigy, publishing his first scientific paper at age 15 and graduating with a PhD from California Institute of Technology at age 20. His incredible work spans science, mathematics, and computing, and he developed Mathematica, Wolfram Alpha, and the powerful computational programming language, Wolfram Language.
“Alongside fundamental science, my main work in life is building the Wolfram Language computational language, which aims to have a way of computationally expressing things that are useful to both humans and computers,” Wolfram told TechCrunch.
As AI developers and others start thinking more deeply about how computers and humans intersect, Wolfram says it will become more of a philosophical exercise that involves thinking in a pure sense about how this kind of technology will affect humanity. That kind of complex thinking has ties to classical philosophy.
“The question is, what do we think about? And that's a different kind of question, one that's more common in traditional philosophy than in traditional STEM,” he said.
For example, when you start discussing how to put guardrails on AI, this becomes an inherently philosophical question. “In the tech industry, when we discuss how to configure this or that with AI, some people say, 'Just make it do the right thing,' and that leads to the question, 'So what is the right thing?'” And determining the moral choice is a philosophical exercise.
He says he's had “scary discussions” with companies putting AI out there who clearly haven't thought about this: “When you have Socratic discussions about how to think about these issues, you're surprised at how clearly people haven't thought about these issues. Right now, I don't know how to solve these problems. That's the challenge, but I think this is where these philosophical questions are important right now.”
Scientists, he says, generally have a hard time thinking philosophically about things. “One of the really striking things I've noticed is that when you talk to scientists and you tell them about big new ideas, they get confused, because that's not what typically happens in science,” he says. “Science is an incremental field, and we don't expect to be confronted with big, different ways of thinking about things.”
If the main task of philosophy is to answer big ontological questions, he believes that the growing influence of AI, and all the questions it raises, will usher in a golden age of philosophy. In his view, many of the questions we now face because of AI are actually at the heart of traditional philosophical questions.
“I find that the group of philosophers I speak to are actually much more agile when it comes to thinking paradigmatically about different kinds of things,” he said.
One such encounter during his travels was with master's of philosophy students at Ralston College in Savannah, Georgia, where Wolfram spoke to the students about a future where the liberal arts and philosophy collide with technology. In fact, Wolfram said he reread Plato's Republic in an attempt to return his thinking to the roots of Western philosophy.
“And this question is: 'If AI is going to run the world, how do we want it to run the world? How do we think about that process? What does the modernization of political philosophy look like in the age of AI?' These are questions that go straight back to the fundamental questions that Plato asked,” he told the students.
Rumi Albert, a student in the Ralston program who is pursuing a career in data science and also participates in Wolfram Summer School, an annual program designed to help students understand Wolfram’s approach to applying science to business ideas, was fascinated by Wolfram’s thinking.
“It's really interesting that someone like Dr. Wolfram has such an interest in philosophy, and I think that speaks to the importance of philosophy and a humanistic approach to life, because it seems to me that he's developed so much in his field that it's evolved into more philosophical questions,” Albert said.
That Wolfram, who has been at the forefront of computer science for half a century, sees the connections between philosophy and technology may signal that it's time to consider these questions about the use of AI in a broader context than simply as mathematical problems — and maybe bringing philosophers into the discussion is a good way to do that.
