We've already discussed how the Israel-Hamas war is the latest conflict where people are poring over social media and news channels looking for updates on what, exactly, is happening. After all, whether it's news about our neighborhoods or communities on the other side of the world, the web is where we go to find updates.
And it's another reminder that misinformation is often big business, and it's everywhere: fake news and fabrications, half-truths and obfuscations, and flat-out lies and propaganda. The rise in AI-powered deep fakes has only made the problem worse and increased the amount of untrustworthy content out there.
So is it actually still possible to filter truth from lies online? We don't yet have a foolproof way of checking—perhaps that's a task AI could be trained on next—but there are ways to limit the likelihood of being fooled.
If you buy something using links in our stories, we may earn a commission. This helps support our journalism. Learn more.
When you buy something using the retail links in this email, we earn a small affiliate commission. Read more about how this works.
This email was sent to you by WIRED. To ensure delivery to your inbox (not bulk or junk folders), please add our email address, wired@newsletters.wired.com, to your address book.
0 Comments:
Post a Comment