Hmm...ourworldindata shows the doubling time as having increased to 10 days in France, new cases declining and the curve flattening, so not sure where this person is getting their data and/or analysis, but it seems clearly wrong.
The article just counts the total number of cases, without mentioning the number of tests, right?
Say on day one you test 10 persons and get 1 positive. Then on day two you tests 100 persons and get 10 positive.
You can say it's 10% positive in both cases: the virus doesn't spread. Or you can say the number of positive was multiplied by 10 in only one day: wow, "exponential growth", the number of positive cases is "soaring".
So, yeah, like most of the mainstream coverage around covid, it's just BS. Put the highest numbers along with a few dramatic words in the headline, and you're done, you wrote an article. Just fill the article with whatever, who cares.
https://ourworldindata.org/coronavirus