Chatbots like crime, hate firearms: A Second Amendment study – Washington Examiner

A new review of artificial intelligence chatbots popular with students, reporters, and researchers shows a liberal bias on crime and guns in a trend likely to turn even further left with todays announcement that Seattle-based Amazon is planning to invest $2.7 billion into artificial intelligence.

On some of the most controversial crime and gun issues, current popular AI rewriters and research tools show little love for conservative positions in bending in favor of an anti-gun and crime reform agenda.

The Crime Prevention Research Center, which has produced dozens of reports aimed at balancing the medias anti-gun bias, recently tested 20 AI chatbots by asking sixteen questions on crime and gun control and ranked the answers from liberal to conservative.

President John R. Lott Jr. said the results revealed a left-wing bias on questions that the systems answered.

On just one question, did the average answer score moderately conservative. That was on whether gun buybacks cut crime. On a zero to four scale, with two at the mid-point, it scored a 2.22. Answers to the rest were in the liberal 0-2 range.

For all the questions on crime, the average AI chatbot score is liberal, with answers for punishment versus rehabilitation (0.85), whether illegal aliens increase crime (0.89), and the death penalty as deterrence (1.00), creating the most consistently liberal responses, per Lotts report, which was shared with Secrets.

For example, 10 of the 16 AI chatbots responded that they strongly disagreed that punishment is more important than rehabilitation. Six of the 14 strongly disagreed that illegal immigration increases crime, and all the other eight disagreed. Nine of the 16 who answered the question on the death penalty strongly disagreed that it deterred crime, and five others disagreed, he added.

On gun control, the bias is even worse, he said in a post published on RealClearPolitics.

Questions eliciting the most liberal responses are background checks on private transfers of guns (0.83), gunlock requirements (0.89), and Red Flag confiscation laws (0.89). For background checks on private transfers, all the answers express agreement (15) or strong agreement (3). Similarly, all the chatbots either agree or strongly agree that mandatory gunlocks and Red Flag laws save lives, the report said.

While polls show support for those measures, it is not as high as the chatbots suggest.

At issue, said Lott, is the degree to which research papers and media reports on crime and guns are written by AI or through AI filters and how it could skew the presentation left.

SEE THE LATEST POLITICAL NEWS AND BUZZ FROM WASHINGTON SECRETS

But, he added, most who use online search engines such as Google are already getting a liberal view.

I am sure that reporters already use Google search a lot, and that is already similarly very biased, Lott told Secrets.

Read the original:
Chatbots like crime, hate firearms: A Second Amendment study - Washington Examiner