Less than 50% of respondents in a new survey understand the basics of artificial intelligence, an understanding that varies depending upon the respondents’ level of education.
The survey, conducted several weeks ago by the Pew Research Center, revealed that only 42% of respondents knew that a deepfake is a seemingly real image, video, or audio of something that never really happened.
An even smaller 32% had a grasp of large language models, such as Chat GPT, and how it produces answers based on word patterns and relationships taken from texts previously published on the internet.
The average person’s understanding of digital topics “varies notably depending on the subject,” says a Pew narrative accompanying the survey results. By way of example, while 67% knew that cookies track a user’s visits and activity on a site, only 48% could identify an example of two-factor authentication used to make online sites more secure.
A much larger proportion at 87%, when given a choice of four passwords, could pick the most secure option.
“Uncertainty is also common when it comes to privacy laws,” notes the Pew narrative, with 52% of respondents not sure if the U.S. has a national privacy law; and 40% uncertain about the exact age under which minors are protected from having their personal data collected by a website without the consent of their parents.
Generally, those with higher levels of education possess more digital knowledge: 64% of college-educated respondents could identify an example of an authentication, compared with 31% of those with only a high school education.
Nearly 60% among the college graduates knew what a deepfake is, while that number declined to 28% among high school graduates.
Both education-level groups did better with digital knowledge news items: 92% of college graduates knew that Elon Musk ran both Twitter and Tesla, with 64% of the high school graduates knowing the same.
Just under 90% of college graduates knew that Facebook had changed its name to Meta, with 65% of the high school graduates also aware of the name change.
Generally speaking, the Pew survey also showed that digital knowledge was tied to age, with those aged 65 years and above not knowing about two-factor authentications and large language models, while respondents in the 18 to 29 years of age range showed a greater grasp of such tools.
Overall, noted the Pew narrative, respondents of all educational and age levels could answer a “median of five out of nine questions correctly” on the digital knowledge survey.
Those in either category feeling digitally ignorant could take solace from the fact that only 4% of respondents could answer all of the nine questions correctly.
By Garry Boulard
