Edwin Chen

Article

Edwin Chen is a recurring person in the Astral Codex Ten archive, appearing 4 times across 4 issues between June 27, 2022 and July 08, 2025. The archive places it in contexts such as “Edwin Chen (how do humans do on the same questions Gary Marcus asked GPT?)”; “Edwin Chen has surveyed some people on whether the Imagen images on my AI art post match the prompts”; “Edwin Chen has tested them out, and says they’re not quite there”. It most often appears alongside Gary Marcus, Scott, Vitor.

Metadata

  • Category: People
  • Mention count: 4
  • Issue count: 4
  • First seen: June 27, 2022
  • Last seen: July 08, 2025

Appears In

Source Context

Recovered passages from the original issue text. When the raw archive preserved outbound links inside the source passage, they are listed directly under the quote.

June 27, 2022 · Original source
2: More volleys in recent AI conversations: Cameron Bucker (“debates in deep learning are now repeating the same mistakes as comparative psychology”), Edwin Chen (how do humans do on the same questions Gary Marcus asked GPT?), and a new paper, Emergent Abilities Of Large Language Models (formalizing the insight that as models scale up, they can do completely new types of tasks, not just the old tasks better).
October 09, 2022 · Original source
2: Edwin Chen has surveyed some people on whether the Imagen images on my AI art post match the prompts, and most people believe that 1-2 do, rather than the 3 I claimed. Given that, I am retracting my claim to have won the bet - which I guess is still on - and adding this to my Mistakes page. Sorry to Vitor and others. I will revisit this sometime in or before 2025, I guess!
January 15, 2024 · Original source
2: Last year I bet that AI art generators would be able to handle some tough compositionality requests before 2025. There was widespread speculation that the latest generation (eg DALL-E2) had won this bet. Edwin Chen has tested them out, and says they’re not quite there.
July 08, 2025 · Original source
In September 2022, I got some good results from Google Imagen and announced I had won the three-year bet in three months. Commenters yelled at me, saying that Imagen still hadn’t gotten them quite right and my victory declaration was premature. The argument blew up enough that Edwin Chen of Surge, an “RLHF and human LLM evaluation platform”, stepped in and asked his professional AI data labelling team. Their verdict was clear: the AI was bad and I was wrong. Rather than embarrass myself further, I agreed to wait out the full length of the bet and re-evaluate in June 2025.
On prediction markets, where users had given 62% probability that Edwin would grant me the win that year, reactions were outraged. “Are you kidding me?” asked one commenter. “Is Edwin Chen an asshole? Clearly he is,” said another.