Do AI tools have 'opinions' and 'political preferences'?

Well, kind of, because they are all based on pre-configured or inferred 'values' which impact their responses, especially when your question is formulated in a specific way.

An insightful study just came out to analyze this topic - here is a brief summary of its main points.

Hope you'll find it helpful.

(More such summaries are coming, so make sure to follow/subscribe.)

*This was originally posted on Andras Baneth's LinkedIn account.