While I’m no expert, that doesn’t sound correct to me. I’d expect highly specific binding dependent on the chemical structure of glucose would be required to elevate insulin. A quick search seems to support that. I’m sure there are lots of studies on this that you could find if interested.
I’m assuming it was small bench-top laboratory covid testing instruments for single samples that reported in minutes. These came after large batch sample testing that took many hours. Where I’m from, early in the pandemic, people had to drive to get rested and results were around 24-48 hours later.
So yes, ‘covid testing machines’. Those aren’t the words I’d choose but fair enough for general reporting.