CNET paused publication of content generated by AI earlier this year. Despite there being some very good reasons for that decision, they now say they’re going forward with it once again.
The reality?
The AI tool was publishing blatantly false and occasionally plagiarized material across CNET’s websites.
Yeah. That’s a problem.
Red Ventures, who owns CNET, is rebranding their AI tech as an assistive tool, and according to VP of Content for Financial Services Lance Davis, things will be different this time around.
“As you all recall, we put a temporary pause on AI-assisted content. I want to talk about a few of the process changes we’re planning to make, both to the tooling as well as to some of our editorial checkpoints that come up in the content creation process.”
Managing Editor of AI-Powered Content, Cameron Hurta, said there are “enhancements” put in place this time around.
“Our primary tooling enhancements kind of go into three different buckets: output, quality controls, and user experience.”
As far as output, Hurta says they have more accurate ways of checking for plagiarism. He also says that new-and-improved machinery will offer “quality and AI likeness scores for all generations.”
Image Credit: iStock
Detractors say the company seems more concerned with whether or not Google’s algorithms will be able to detect the AI-generated content than user experience.
“The likeness scores will help us ensure that our content still reads human-like and is not going to get tagged as being AI content. And that, you know, the generations you’re getting are of solid quality.”
Red Ventures has been cagey in answering questions about whether or not AI-generated content would be marked going forward. Likewise, people are skeptical of their claims that the new “tooling” system will be able to add source URL’s into their generated content, as no other AI tool has managed to learn that bit as of yet.
People are also dubious when it comes to the claim that “editorial checkpoints” managed by an “expert review board” will make much difference as far as fact checking, either, since they claimed previously that all AI-generated content was already being double-checked by human editors.
In the end, we’ll all just have to wait and see whether or not CNET was actually able to turn their AI issues around this quickly, or if more trouble is lurking on the horizon.
Image Credit: iStock
Given all of the troubles we’ve seen others have, it feels as if this is a hasty move, but only time will tell.
And everyone knows at this point that the first company to have an AI tool that will tell the truth and not be insanely creepy sometimes is basically going to win.
What the prize is, well…we’ll just have to wait and see.