This may be old news to some of you, but it is not to me: Feedly, once a RSS feed reader, is an AI-driven threat intelligence company now. Apparently this has been in the works for some time. As of January, the Feedly website advertised the company as three things—threat intelligence, market intelligence, and a news reader. Months later, Feedly all but scrubbed market intelligence and its news reader from its website. (A small button, "Go to News Reader," remains on the top of the front page.) I suppose this is news to me because I never actually go to the Feedly website; I just go straight to my RSS feed.
Feedly has been kind of bad lately, which is how I found out that it made the baffling pivot to AI: My partner mentioned it after I complained about how much Feedly sucks, and I then found a news story from TechCrunch in which Feedly claimed its summer slowdown was a bug and not because of AI. In the last year, Feedly Classic stopped working. The Feedly app is broken. It seems to be slowly disabling its news reader features. The company had seemingly been doing nothing to address this stuff. (Some of this stuff has been fixed, apparently. But Feedly still sucks for me.)
“Despite the pivot to cyber threat intelligence, fixing the basic RSS functionality is still a priority because our CTI community uses the news-reading capability,” Feedly CEO Edwin Khodabakchian told TechCrunch. “Our goal has been to maintain the Feedly News Reader as fast and streamlined as possible.” Sure!
I understand that this is a niche issue; overall, I think there's a very small population of people who still use RSS feeds. But they’re helpful for me as someone who reads a lot of news, since it’s near impossible to keep up with everything as a journalist without some sort of collation. Feedly has never been a good option, but there are very few options, and nothing ever really hit like Google Reader did.
I can't even really tell what Feedly is supposed to do as a threat intelligence product. It actually looks a lot like news reader Feedly. Apparently you can use its AI agents to look at the dark web for security vulnerabilities? "Deliver STIX-formatted threat intel to your entire security stack and stakeholders?" What the fuck?
It's expensive, too. It starts at $1,600 per month and goes up to $2,400 monthly. I simply don't understand. It's the sort of mind boggling pivot that I attribute to Allbirds selling off its shoe business to do something with AI.
I don't know, guys. I don't think I'd trust AI to help me with internet security. There's a whole Wikipedia page about cyberattacks from rogue AI agents! Last month, AI exploited a bug that created a big ol' crisis with a PHP vulnerability. A security research company did find the bug in this case, but the ability of AI to very quickly do these sorts of things makes it much easier for people to use it to create chaos, steal information, and generally wreck havoc on the internet.
