Google's powerful AI spotlights a human cognitive glitch: Mistaking fluent speech for fluent thought

By Kyle Mahowald and Anna A Ivanova

The requested article has expired, and is no longer available. Any related articles, and user comments are shown below.

© The Conversation

©2022 GPlusMedia Inc.

Login to comment

I'll post this again as ,any AI theorists posit an exactly human model of cognition may not be necessary.

The question of whether a computer can think is no more interesting than the question of whether a submarine can swim.

Edsger W. Dijkstra

1 ( +1 / -0 )

If there is a true AI then that submarine can swim.

0 ( +0 / -0 )

One of the most interesting things of these kind of AI tests is how it leads to a lot of better understanding of how the humans react and interpret the output received. With an AI trained to fool humans into thinking the language comes from an rational though process it becomes quite clear what kind of cheats and shortcuts are more effective for this purpose. Maybe a sentient AI is still long in the future, but in the meanwhile we can learn a lot about why our human brains can be fooled so easily.

-1 ( +0 / -1 )

Peanut butter, jelly, pineapples and feathers are fine. I will start getting worried when an AI says - I’ll be back.

2 ( +2 / -0 )

Login to leave a comment

Facebook users

Use your Facebook account to login or register with JapanToday. By doing so, you will also receive an email inviting you to receive our news alerts.

Facebook Connect

Login with your JapanToday account

User registration

Articles, Offers & Useful Resources

A mix of what's trending on our other sites