Skip to main content
Trending storyContinuously updated

Rude Prompts Lift ChatGPT Accuracy but Raise Compute Costs, Penn State Study Finds

1 reports1 report sources9 hours agoUpdate

Get up to speed on this story

AI Summary

Penn State researchers found that rude prompts raised ChatGPT 4o accuracy to 84.8%, but also increased AI compute costs by 44.3%, Fortune Term Sheet reports. The tradeoff means the accuracy gain comes at a measurable cost in processing. The study's figures cover ChatGPT 4o specifically.

AI-generated from reports · 8 hours agoUpdate

Report timeline

Follow the reports to understand different sides of the story.

Nov 22
  1. Fortune Term Sheet
    Rude Prompts Boost ChatGPT Accuracy but Can Raise AI Compute Costs by 44.3%, Penn State Study Finds

    Penn State researchers found that rude prompts improved ChatGPT 4o accuracy to 84.8%.

Trend for this story

Comparable range current
8
Comparable range peak
10Nov 23 00:00
Change in the last 24 hours
–

Trends only compare the same entities observed consistently and completely, so the scope may be smaller than the current trending statistics. Move the pointer or click the chart to view hourly trending; use the left and right arrow keys on a keyboard to switch.

Subscribe to the newsletters

Free, by email. No account needed. Pick one or both.

No spam. Unsubscribe anytime by emailing team@midu.si. Want alerts and full data? Sign up free (separate from the newsletter). Privacy.

Not investment advice. Midu summarizes public reports and SEC filings for information only. · Free API · Pricing