That's because that's how science works. Discoveries are not considered to be statistically significant until they reach what's called 5 sigma certainty which is approximately equivalent to saying that the chance that the discovery is wrong is 1 in 3.5 million.
A lot of scientists would consider it unethical to claim a discover until you had provided enough data to reach 5 sigma certainty. When papers are published, it takes a lot of peer review before the hypothesis of that paper event approaches 5 sigma certainty, but that doesn't mean that reporters aren't happy to pick up the story.
It's just bad and/or unethical science journalism that are picking up on unproven papers because of the sensational title.