Gary Marcus

Rank 14 of 47
|
Score 83
Gary Marcus
@GaryMarcus
·
636d
So many people are confused about the relation between human cognitive errors and LLM hallucinations that I wrote this short explainer:

Humans say things that aren't true for many different reasons
• Sometimes they lie
• Sometimes they misremember things
• Sometimes they fail…
5/15/2024, 10:12:46 PM
X
In reply to:
Sean McClure
@sean_a_mcclure
·
611d
@GaryMarcus @xuancanh_dev Should we run a tally of human mistakes versus AI mistakes? Or is that too inconvenient?
Gary Marcus
@GaryMarcus
·
612d
@xuancanh_dev Sometimes. Elsewhere in the thread or QTs someone noted a 4o fail.
Canh Nguyen
@xuancanh_dev
·
612d
@GaryMarcus Interestingly, GPT 4o solves this question correctly.
Gary Marcus
@GaryMarcus
·
612d
Someday my children will ask me why anyone ever took seriously the thought that systems like this had anything to do with AGI.

I honestly have no idea how I will answer. Mass delusion?

(Example emailed to me by a friend.)

The statement '@sean_a_mcclure @xuancanh_dev' does not constitute public discourse as it lacks substantive engagement with public issues, debates, or discussions that contribute to the civic dialogue. It appears to be a part of a conversation thread on social media, but without additional context or content in the statement itself, it does not address societal issues or policies.