My first technological revolution was internet as a commodity. The AOL dial-up sound, raising my Neopets, and setting angsty song lyrics as my status in chat rooms— it all feels strangely so close and far. If you’re a current college student, these early days of GPT may end up being similar memories for you. Though as one person tweeted, “This feels less like an ‘iPhone’ moment and more like ‘fire’“.

What entry-level jobs are vulnerable?

In this paper co-written by OpenAI and University of Pennsylvania, vulnerability to AI is defined as “driving a reduction in time it takes to complete the task by at least 50%”. Here are the most exposed jobs, according to the paper:

• Mathematician
• Tax preparers
• Financial Quantitative Analysts
• Writers & Authors
• Accountants & Auditors
• News, Analysts, Reporters, and Journalists
• Legal Secretaries and Administrative Assistants
• Clinical Data Managers
• Climate Change Policy Analysts
• Web and Digital Interface Designers

The paper further breaks down skills that are most vulnerable, which correlate strongly to the above list:

• Math
• Programming
• Writing
• Listening

On the other end of the spectrum, the skills that are the least vulnerable:

• Reading Comprehension
• Science
• Learning Strategies
• Monitoring
• Crticial Thinking

So does my Computer Science degree even matter anymore?

GPT does not change my advice on this matter. You should major in something you find interesting and curious, regardless of how you think it translates into a job. The paper says programming will become very efficient, but there will still always be a need for programmers. More specifically, there will be an even greater need for programmers that have reading comprehension, know how to learn new things, and can think critically about a problem space.

If you are pursuing something out of passion and intrinsic interest, it will be a lot easier for you to develop those soft skills. Soft skills is a bit of a misnomer because they are clearly the harder skills to develop than the pure technical.

As you’re going through your classes, think critically about the tools you are using and why. If you’re wondering if you should use GPT to write code, I say go for it, sure. It will clearly become an industry tool and becoming familiar with it will be like knowing git or docker. But, as you’re using GPT, remember that you as the human bring all of the reading comprehension and scientific process. Understanding what you’re doing is just as important as getting the task done!

Are you doing something boring, manual, and necessary with your content? Let us know - we’d love to help.