Google was previously the undisputed intermediary of the Internet. Users received the best possible links to Internet pages for their search queries. What was on these pages, whether it was true and up to date, was outside of Google’s responsibility. If Google provided direct answers, it was only for objective facts such as the current time in New York or Tom Cruise’s height.
Now, Google is testing AI Overviews in the USA – summaries of content generated by artificial intelligence.
The media response to this has, rightly, been devastating. For example, when asked about depression, Google recommended jumping off the Golden Gate Bridge, mixing glue into cheese to make better pizza sauce, and advise for pregnant women to smoke two to three cigarettes a day.
The fact that the Internet is full of untruths, lies and manipulative content is nothing new. What is new is that Google is now publishing this content itself in the search results.
Until now, the search engine enjoyed the privilege of being only an intermediary. With the introduction of AI Overviews, Google has given up this advantage. Users perceive this content as coming from Google and expect it to be correct. How Google intends to solve this problem is currently completely unclear.
