Scott Aaronson in Shtetl-Optimized:
As I’ve mentioned before, economist, blogger, and friend Bryan Caplan was unimpressed when ChatGPT got merely a D on his Labor Economics midterm. So on Bryan’s blog, appropriately named “Bet On It,” he made a public bet that no AI would score on A on his exam before January 30, 2029. GPT-4 then scored an A a mere three months later (!!!), leading to what Bryan agrees will likely be one of the first public bets he’ll ever have to concede (he hasn’t yet “formally” conceded, but only because of technicalities in how the bet was structured). Bryan has now joined the ranks of the GPT believers, writing
When the answers change, I change my mind
AI enthusiasts have cried wolf for decades. GPT-4 is the wolf. I’ve seen it with my own eyes.
But OK, labor econ is one thing. What about a truly unfakeable test of true intelligence? Like, y’know, a quantum computing test?
Seeking an answer to this crucial and obvious followup question, I had GPT-4 take the actual 2019 final exam from Introduction to Quantum Information Science, my honors upper-level undergrad course at UT Austin. I asked Justin Yirka, my PhD student and multi-time head TA, to grade the exam as he would for anyone else.