Goldman Sachs CIO Tells Engineers to Study Philosophy

Learning how to code? A standard engineering degree might not be enough.

In a recent post in the Harvard Business Review, Goldman Sachs Chief Information Officer Marco Argenti said that AI can now write high-quality code, sometimes greater than humans, and the technology is only getting better.

So how does an engineer keep up? He told readers the advice he gave his college-age daughter: “If you want to pursue a career in engineering, you should focus on learning philosophy in addition to traditional engineering coursework.”

Argenti is an engineer with a background in tech entrepreneurship and leadership at other large companies like Amazon Web Services and Nokia.

^{Marco Argenti, CIO of Goldman Sachs, on the left, and Robert Kyncl, CEO of Warner Music Group, on the right. Credit: Patrick McMullan/PMC via Getty Images}

He said that run-of-the-mill engineering coursework isn’t enough as AI expands in complexity becasue AI code can be technically correct but not do what it’s supposed to do.

If an engineer doesn’t get the prompt right, the AI will create code that could be somewhat right in the best-case scenario and outright incorrect in the worst-case scenario. But engineers grounded in philosophy should have the reasoning abilities and first-principle thinking to keep pace.

The question for engineers will not be, “Can you code?” but “Can you get the best code out of your AI by asking the right question?”

Related: What Is Prompt Engineering, the Hot New Tech Job in AI

AI prompt engineering, or communicating with AI in the best words to get the desired output, is an emerging field. Jaime Teevan, chief scientist at Microsoft, wrote in December that communicating with AI is different from telling a team member what to do. AI needs more context and might need rephrasing or examples before it produces a workable output.

PricewaterhouseCoopers partner Robin Stein told the publication CIO that PwC “realized very quickly we had to upskill our 75,000 people on the foundations of generative AI, how to apply gen AI responsibly, and how to become a prompt engineer.”

Related: Two Yale PhDs Are Trying to Make AI Hallucinate 10x Less, So Small Businesses Can Adopt the Technology in Weeks, Not Years

Argenti emphasized that in an era of AI hallucinations, it was all the more important to ask the right questions and employ a philosophical mindset.

“Having a crisp mental model around a problem, being able to break it down into steps that are tractable, perfect first-principle thinking, sometimes being prepared (and able to) debate a stubborn AI — these are the skills that will make a great engineer in the future, and likely the same consideration applies to many job categories,” Argenti wrote.