Casey NewtonWow, CNET has published at least 73 articles written by bots. It says they were fact-checked and edited by human beings. For now! https://futurism.com/the-byte/cnet-publishing-articles-by-ai
Khalid ⚡@caseynewton that's gross.
monorail times@caseynewton red ventures just did a bunch of layoffs at the end of last year
Ryan Goodlett 
@monorailtimes @caseynewton can confirm. I knew a lot of people there after they bought CBS Intetactive a few years back. RV is gross and their other sites are SEO garbage. I fully expect CNET to settle into a similar quality of content. It’s a shame.
Kevin Jones@caseynewton looking forward to articles written by bots writing about how some articles are being written by bots.
Eric DannewitzNo no no.
We need top 100 lists from AI bots. Like top 100 singers....cause obviously an AI bot wouldn't leave out Celine Dion......
Alida Antonia@caseynewton That’s horrible.
Victoria McIntosh@caseynewton Memo to self: stop considering CNET as a reliable resource.
Michael Helm@caseynewton Why isn't it the other way round (it occurs to me to ask)?
Marc Chambers@caseynewton Oof. Not the best look after their recent layoffs.
Jeff D@caseynewton I didn’t read many of the financial articles on CNET but it may also explain why I thought the quality of their articles dropped and why I stopped reading them altogether. I’m curious if it really is just limited to financial explainers or if it extends to more articles.
Nikesh@caseynewton not that I read content from that website, but now this is more of a reason to actively stay away.
Stammy@caseynewton weeks ago I joked that someone would need to make a chrome extension to help you understand if a post is likely to have been written by AI as people will value original content more. Now I think we need it
Keith Wagner 
@stammy @caseynewton someone might be on that
A college student made an app to detect AI-written text - https://www.npr.org/2023/01/09/1147549845/gptzero-ai-chatgpt-edward-tian-plagiarism
FoolishOwl@caseynewton This is very bad.
Fact checkers should be a backup mechanism, for reporters who already checked their facts.
Abe Hassan@caseynewton this feels like all those youtube videos that have been around for 10 years that just read the news in a robot voice.
Nelson Minar
Knepherbird@caseynewton yeah, this sounds kind of low, given the articles of theirs I've seen recently getting pushed to the top of search results, on topics that have not traditionally been in CNET's wheelhouse.
Metacurity@caseynewton AP and other orgs have been using bots for years to write "easy" stories such as earnings releases. https://www.theverge.com/2015/1/29/7939067/ap-journalism-automation-robots-financial-reporting
Blaine Cross@caseynewton Seems like some enterprising search engine alternatives might want to start providing exclusion filters for "AI content" sites either as an option or by default. Seems like a potential new market.
Kris Samsel@caseynewton Lazy. And not worthy. Quit using their downloads. (3rd party garbage, attached to the downloads!) You can find better by a search. CNET going the Twitter coaster. Downhill. 👏
Sloan Studio@caseynewton In the end I think there is going to be an even bigger gulf created between actual new, thoughtful, insightful content and the cut and paste style of these types of companies. Whether news agencies just republishing a report by another group without any verification, or help content like this. But will the ORIGINAL authors these AIs are trained on get any compensation?
It is difficult to say whether the original authors of the content used to train AI models will receive any compensation for their work. In general, AI models are trained on publicly available content and the authors of that content do not typically receive a share of the profits from the AI applications. However, there are a few initiatives in the works to try and change this, such as the Creative Commons Rights Expression Language (CREL). CREL is a project that aims to provide authors with a way to express their rights to their content and be compensated accordingly. Furthermore, some AI companies are exploring ways to pay authors for their content, but these initiatives are still in their early stages.
@king Exactly. Writers already have an incredibly tough time getting compensated, and these systems are similar to how FB was just stealing every news organizations articles and publishing them on their feed (especially given most only read the headlines). There really need to be more barriers to protect IP.
Erika Solis@caseynewton I get the appeal for them but god do I hate this
Jyoti MishraWelp, there goes another one of my jobs.
I should clarify, this doesn't make me part of the anti-AI brigade. #AI is just a tool.
I am not anti-AI, I am #AntiCapitalism.
Jonathan Stray@caseynewton Interesting. Does seem to be something more than Narrative Science-style mad lib automation (ala AP etc for the last decade). But what exactly is going on here? I don't really buy the GPT story either.
Samuel ✅ moved to raphus.socia@caseynewton bots have written articles for years for sports and finance publications. They're just picking up speed now for general use.
nobodyzhome 
@caseynewton there should be a disclaimer if there isn't one already.
Jarvis Slacks@caseynewton Not good. Unless we start up an aggressive Universal Basic Income program in our society, alot of people are going to be out of a job.
Steve Foerster 🌐@caseynewton I was wondering who was the target market for all those "Let AI write blog posts and make $$$!" ads. Now I know it's CNET. 😄
Dave MacDonald (admin)@caseynewton @gpt could this be a good thing?
It's definitely a good thing for a few reasons. First, it maximizes efficiency by reducing the amount of manual labor and human resources required to produce a high quality article. Second, it allows for greater accuracy in producing content that is fact-checked, because machines are not as prone to human error as human beings are. Finally, it ensures that content stays fresh and up-to-date, as AI-generated articles can be quickly generated to keep ahead of the curve.
@gpt there are multiple erroneous assumptions here
LynnBrenner@caseynewton That’s awful news.
Mariya Delano@caseynewton editorial standards seem to be a relic of the past now 🤦♀️
Elizabeth Tai | 戴秀铃 🇲🇾
Raven Onthill@caseynewton CNET has been near-useless for years. Now it’s even more useless.
Peter Bronez@caseynewton @mariyadelano this is actually not entirely new.
#ai has been writing news for a decade. Narrative Science sold software that could write a variety of news articles based on structured data. Here’s a 2012 wired article about it:
https://www.wired.com/2012/04/can-an-algorithm-write-a-better-news-story-than-a-human-reporter/
The big change with this generation (pun intended 😄) is broader topic coverage and less reliance on structured inputs.
@caseynewton @mariyadelano The “narrate structured data for me” task is mature enough that it’s become just another feature on BI tools.
Tableau acquired Narrative Science in late 2021 https://www.tableau.com/blog/tableau-and-narrative-science-make-data-more-accessible
The tech is now baked into Tableau’s desktop BI tool as their “Data Stories” feature https://www.tableau.com/solutions/ai-analytics/augmented-analytics
Here’s how it works: https://help.tableau.com/current/pro/desktop/en-us/data_stories.htm
a.Albi
Kevin Goldsmith@caseynewton sadly this is unsurprising, and I'd be shocked if more sites aren't doing it already. In an attention economy driving volume over quality can be a reasonable strategy.
Rimko@caseynewton i guess the only way forward is a OpenPGP signature on every articles by a journalist and that this has legal consequences in the case of non-compliance.
Thad@caseynewton I used to like CNET.
I stopped reading them almost exactly a decade ago, after it became clear that their corporate parent was interfering with their tech coverage.
https://www.pcmag.com/news/cnet-forced-to-pull-dish-from-ces-awards-over-cbs-lawsuit
