It absolutely can be worse. Right now, their limited subset of answers would be wrong in a way that is easily correctable. They fail safe. An LLM doesn’t fail

It absolutely can be worse. Right now, their limited subset of answers would be wrong in a way that is easily correctable. They fail safe. An LLM doesn’t fail safe. It fails confidently. Correcting bad responses can be a very expensive and time consuming effort. In the meantime you get actively harmful suggestions: https://arstechnica.com/information-technology/2024/05/googles-ai-overview-can-give-false-misleading-and-dangerous-answers/ https://arstechnica.com/information-technology/2024/05/googl... Alexa telling you the wrong weather is a lot less harmful than it telling you to do something that would harm you (staring at the sun, running with scissors, smoking when pregnant) or spreading political misinformation (telling you conspiracy theories).