/AI6h ago

AI Code Productivity Metrics Decline Sharply From Feb To Apr 2026

428122.8K
#225Christopher Potts
Original post
Christopher Potts@ChrisGPotts#225inAI

To give you a concrete sense for what the above declines mean, here are the estimated outcomes for a 10K-token session:

Feb 2026: a third of a PR, 630 lines of code drafted, 8 files touched, 1 knowledge capture.

Apr 2026: a sixth of a PR, 91 lines of code drafted, 4 files touched, less than half a knowledge capture.

Christopher Potts@ChrisGPotts

Our CPI includes a hedonic adjustment based on the code survival rate. This goes from about 90% to 95% in our time period, so we incorporate a uniform 5% adjustment. This softens the tokenflation; code output seems to be improving!

9:24 AM · Jun 8, 2026 · 677 Views
Sentiment
Sentiment building, check back later.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS589
Christopher Potts@ChrisGPotts

We would love to expand our analysis to other models, providers, and time periods. We've created a Claude Code tool/plugin that you can use to analyze your own usage. The tool shares no data with us, but perhaps we could all pool our anonymized data to learn more about the true value of our tokens: https://github.com/bigspinai/plugins

Christopher Potts@ChrisGPotts

We ourselves derive significant value from these tools and wonder whether measuring other relevant outcomes would change the CPI. That said, it seems clear that AI is becoming more expensive in ways that might be outpacing its current value.

6hViews 589Likes 5Bookmarks 1
BOOKMARKS1REPLIES1
Christopher Potts@ChrisGPotts

We ourselves derive significant value from these tools and wonder whether measuring other relevant outcomes would change the CPI. That said, it seems clear that AI is becoming more expensive in ways that might be outpacing its current value.

Christopher Potts@ChrisGPotts

What is causing the decline in token-level purchasing power? One relevant thing we know is that the job of an Opus 4.6 token changed dramatically in this period, from purely writing code in February to coding, thinking, and explaining in mid-April:

6hViews 465Likes 6Bookmarks 1
LIKES7
Christopher Potts@ChrisGPotts

What is causing the decline in token-level purchasing power? One relevant thing we know is that the job of an Opus 4.6 token changed dramatically in this period, from purely writing code in February to coding, thinking, and explaining in mid-April:

Christopher Potts@ChrisGPotts

To give you a concrete sense for what the above declines mean, here are the estimated outcomes for a 10K-token session:

Feb 2026: a third of a PR, 630 lines of code drafted, 8 files touched, 1 knowledge capture.

Apr 2026: a sixth of a PR, 91 lines of code drafted, 4 files touched, less than half a knowledge capture.

6hViews 551Likes 7Bookmarks 0
Christopher Potts@ChrisGPotts

Another link to the associated blog post: https://bigspin.ai/resources/the-decline-of-token-level-purchasing-power

Christopher Potts@ChrisGPotts

We would love to expand our analysis to other models, providers, and time periods. We've created a Claude Code tool/plugin that you can use to analyze your own usage. The tool shares no data with us, but perhaps we could all pool our anonymized data to learn more about the true value of our tokens: https://github.com/bigspinai/plugins

6hViews 497Likes 5Bookmarks 0