Experts say when ChatGPT confabulates, it is reaching for information that is absent from its training data and filling in blanks with plausible-sounding words
A look inside the hallucinating artificial minds of the famous text prediction bots. — Over the past few months …
Ars Technica Benj Edwards
Related Coverage
- We need to tell people ChatGPT will lie to them, not debate linguistics Simon Willison's Weblog
- ChatGPT's Golden Rule In the future … · David G.W. Birch
- Most INSANE things ever done on ChatGPT (GPT-4) gHacks Technology News · Onur Demirkol
- Got-It AI Debuts Compact Large Language Model With Hallucination Filters ELMAR Voicebot.ai · Eric Hal Schwartz
- ChatGPT 5 set to launch soon boasting Human-like intelligence Gizchina · Abdullah
- Google Search to get ChatGPT-like AI, and I expect Google to fix 2 big problems BGR · Chris Smith
- Here's an accessible piece explaining how ChatGPT works. If you're seeking a quick understanding, basically takes the factuality of ChatGPT's output with a large pinch of salt. … Calvin Chan
Discussion
-
@benjedwards
Benj Edwards
on x
Right now, AI chatbots are the ultimate bullsh*t machines, easily inventing histories, citations, and biographical details that don't exist Why is that, and is there anything researchers can do to fix it? I asked several experts about it for Ars: https://arstechnica.com/... https…
-
@simonw
Simon Willison
on x
Wrote about why I think it's better to tell people “ChatGPT will lie to you”, despite “lying” misleadingly implying intent and the risk of encouraging anthropomorphization https://simonwillison.net/...
-
@smpl5
Martin Weiss
on x
This is a very nicely written piece on how AI Chatbots work and why they're prone to making things up: https://arstechnica.com/... @arstechnica #ChatGPT #chatbot #chatbots
-
@mmitchell_ai
@mmitchell_ai
on x
Let's talk about factuality with Chatbots like ChatGPT! Thanks to @benjedwards for covering details of this super important topic, and for interviewing me. =) https://arstechnica.com/...
-
@larerbloggen
Martin Johannessen
on x
This article explores the issues of AI chatbots generating misleading or false information, and discusses various approaches researchers and developers are taking to improve their accuracy and reliability. https://arstechnica.com/...
-
@cherthedev
Cher Scarlett
on x
I appreciate @benjedwards, but this is another confabulating abstraction that humanizes bots. AI can do what it's programmed to do. Chatbots are generating content from statistically relevant patterns. No magic. No brilliance. The BSing is just us. https://arstechnica.com/...
-
@iethics
@iethics
on x
“Why do #AI #chatbots make things up, and will we ever be able to fully trust their output?” - “There are deeper things one can do so that #ChatGPT and similar are more factual from the start”: https://arstechnica.com/... #ethics #data #tech #business #research
-
@martinfowler
Martin Fowler
on x
🎯 “These are incredibly powerful tools. They are far harder to use effectively than they first appear. Invest the effort, but approach with caution: we accidentally invented computers that can lie to us and we can't figure out how to make them stop” https://simonwillison.net/...
-
@fabiochiusi
Fabio Chiusi
on x
“All this leads to one conclusion, one that OpenAI agrees with: ChatGPT as it is currently designed is not a reliable source of factual information and cannot be trusted as such” Truly a must read article: https://twitter.com/...