I don’t think I’ve ever linked a story from Wirecutter, the NYT’s version of “Consumer Reports” because it’s mostly useful but mundane stuff: “Which is the Best Toothbrush”, stuff like that.
Here’s a piece about how the Lords of AI have wrecked their smart speakers, and it’s kinda funny, so laugh while you can because tomorrow we all die.
Gen AI Was Supposed to Save Smart Speakers. Instead, It Broke Them.
If I want to be ignored, there are plenty of humans I can turn to. I don’t expect it from a device whose entire reason for existing is to listen to me.But that’s exactly what happened when I tested the new **Google Home Speaker**with Gemini AI. It ignored my song requests, changed the subject of conversations, insisted the day was Tuesday on a Wednesday, and then refused to turn off my smart lights, claiming it was “unable to control or interact with your home’s lighting or other physical devices.” Odd flex, since controlling smart devices is kind of the bare minimum one would expect from a smart speaker.
The speaker comes with a six-month trial of Google Home Premium Standard, a subscription plan that adds Gemini AI, which Google says will make conversations more natural and smart-home commands easier to create and carry out. So far it isn’t living up to those claims.
If this kind of thing were a problem with just one device or one AI platform, it would be a disappointment. But it isn’t an individual-device problem. And it isn’t merely a Google problem, either. Whether you’re dealing with Google’s Gemini or Amazon’s Alexa+, the story is the same: Big tech, in its frenzy to make voice assistants smarter and more powerful, has managed to make them worse, by sabotaging their ability to perform the sorts of basic tasks they had evolved to do pretty well.
And the worst part is, these companies now expect buyers to pay extra for the privilege of troubleshooting their half-finished systems.