About this page

GPT-4’s Successes, and GPT-4’s Failures – Communications of the ACM

https://cacm.acm.org/blogcacm/gpt-4s-successes-and-gpt-4s-failures

“GPT-4 is amazing, and GPT-4 is a failure. GPT is legitimately amazing. It can see (though we don’t have a lot of details on that yet); it does astonishingly well on a whole bunch of standardized tests, like LSATs, GREs, and SATs. It has also already been adopted in a bunch of commercial systems...” (from the page’s text)

In short: Explore the successes and failures of GPT-4 in AI, learn about its limitations in truthfulness, reliability, and current events, and understand why it's not yet a solution for chat-style search. (AI summary)

Topic: News Current Events › Technology News
Quality: Quality 84 scored 11 Aug 2025
Language: english
Text on page: 3,806 characters
Page size: 127 kB
Answered: OK (200), HTML
Last read: 7 Jul 2025
In our index since: 7 Jul 2025
Safe search: Not checked yet
Links to it: No other site we've read links here yet

Tools for this page

Other sites that can tell you more. Plain links: we don't track where you go.

About the site

Domain owner ICANN registration lookup
Certificates certificate transparency, subdomains
Safety check Safe Browsing status
Trackers The Markup's Blacklight
Is it down? for everyone, or just you

Discussions

Hacker News stories linking here
Reddit posts from this site
Wikipedia articles mentioning the site

For web makers

PageSpeed Lighthouse report
HTML validator W3C
Accessibility WAVE by WebAIM
Security headers

Who links to it

We haven't seen any links to this page yet. That usually means we haven't crawled the pages that link to it, not that there are none.

“Not rated yet” and similar notes are shown on purpose: we say what we haven't measured, so the page doesn't look emptier or better than it is.