Ai is Everywhere nowadaysBut if you see an article online from a historically a trustworthy source, it is reasonable to believe that a AI is not involved. In the most important newsrooms across the country, however, this reality is becoming increasingly clear.
While not every use of AI in the news editorial team is as obvious as a AI that only designed a contribution to an empty page (although That happens too,) It is important to know how the sausage is created when it comes to the information that shape its worldviews.
Here are just a few news agencies that openly used the AI in their processes, so that you will stay up to date with the precisely the information from which you read.
The New York Times
Early this week, Traffic light Published an article about new AI tools that report the New York Times Management reports to the employees. While the right arm of the publication is currently involved in A Lawsuit against Openaai and Microsoft For copyright infringement, it seems that the website’s editorial and product wings are ready to use AI in the news editorial team, at least through communication from above.
According to Semafor’s article, NYT Management now provides AI training to its journalists, debut an internal AI tool called “Echo” and approves the use of external AI tools, including the Vertex -KI products from Google, some Amazon AI products, And ironically microsoft-Kopilot and ironically a non-chatt tool from Openai.
Not all of this is used for the articles of the website -the New York Times is greater than what you see on the front page -but Semafor said that journalists are encouraged to give these AI tools for tasks such as mild content or mild content revisions To use questions about questions during the interviews.
“Generative AI can help our journalists uncover the truth and help more people to understand the world,” read the company’s company Editorial guidelines on AIthat are public on his website. “We do not consider the technology as a magical solution, but as a powerful tool.”
For this purpose there are apparently still some guardrails. According to the Semafor, the NYT has warned the employees not to use AI to design or revise articles, and found that the AC use may violate copyright or obviously expose sources.
Since internal communication indicates that authors use AI to maintain headlines and to design social copies, it is worth keeping it hallucination Remember the next time you become a suspicious NYT story on social media.
Semafor reports part of the journalists that some remain skeptical and fear that the AI could inspire the “laziness” news editorial team and to reduce the accuracy or creativity.
quartz
First of all, a disclaimer. Quartz is currently owned by G/O Media, which also owned Lifehacker in 2023 before selling Davis.
Nevertheless, G/O media have become one since then Main representative of the AI in the news editorial teamSince Business News Outlet Quartz is its biggest experiment.
Scroll through quartz ‘bylines for a bit and you will find posts that are attributed to this Quartz Intelligence NewsroomWhich apparently drops the restrictions adopted by the New York Times. Here the AI ”Author” has tacitly generated earnings reports for months and recently spit out more general blogs. This includes stories about Potential Bitcoin value or How to delete your meta social media accountsBut as you may expect, there are problems that are aware of everyone.
For Quartz ‘honor it does not hide the fact that these stories are generated with AI and the AI quotes their sources, but it does not seem as if there are human supervision that deals with problems that could result.
For example, the quartz article for deleting your social media accounts in meta-owners Cribdes Techcrunch Writer Calls “vague” gestures in the right direction. The same writer also said about other stories written by Quartz Ai, and also said: “My editor would never let me publish something so sloppy.”
It is very likely that you may find this story on useful advice, just to let it disappoint and confuse than when you came. But the chances that this will only get higher if you end up in a less reliable source in a different story than Techcrunch.
As already mentioned futurismThe Intelligence Newsroom has often cited a website called Devdiscourse, which itself has all the phenomena of a AI content farm. If robots quote robot, I have to ask: Why not just go to chat and demand yourself? Even if I accept Ki messages as worthwhile, I am unclear what efforts are added by G/O here.
G/O strikes a liability exclusion in all quartz stories of ai-generated quartz, from which it can be seen Back to 2023It is unclear whether it will ever find out what this experiment should result.
In the meantime, keep an eye on all quartz bylines, and if you sniff the AI, you should consider the sources to give the sources that the bot combines a reading instead. If you click this attractive heading, you basically play a roulette game.
AP
When the New York Times has just started to dive into the AI and quartz made a complete cannon ball, the use of AI by the AP seems to be somewhere in between. The agency on his website explains proudly That it uses AI for translation, transcription, headlines, research and even some automated articles, but general blogs still remain human hands.
“Our goal is to give people a great way to understand how we can experiment a little, but can also be safe” first turned out his guidelines on artificial intelligence.
If you most likely affect you, the use of Wordsmiths by the website, a AI program that specializes in the summary of content such as Sport -Scores, weather reports and, as with Quartz, has specialized. AP has used a version of this program Since 2014So it’s not exactly new, but it is worth being aware that stories on these topics come from a bot without a specific author. Apart from these stories, the only other clear use of content written directly to AI was an experiment with reporting on public security incidents in A Specific newspaper Minnesota.
Otherwise, the only time is that you will probably see direct AI content in an AP story that summarizes the story in which the pilot summarizes: For example, AI could set a blurb under Election Day stories to say which offices can be obtained.
Here too, headlines and research are still admittedly supported AI, so it is important to be particularly careful if something does not exist, but it seems as if people are still at the fore at least for the time being.
The Washington Post
The use of AI by Washington Post is unique because it does not directly affect the content. Rather, it is more of an improved search engine for readers who are looking for it. The bot called “Ask the Post Ai“Questions”, spits a brief answer from AI-generated, which was trained on the content of the Washington Post, and then lists relevant items below in Google-like.
“Answers are generated from the published reporting,” warns the Bot when you ask a question. “Please check by consulting the items provided.”
The answers seem to aim at a measured tone that is based on quotes from remarkable sources in opinion matters, and when I asked the development of President Trump’s developing attitude to Tiktok or Elon Musk’s story with Tesla, I got truthful answers, although The former was, even though he was the former, even though he was a heel long, while the latter was only one sentence long. I assume how detailed your answers are, depends on the extent to which the paper has treated a specific topic.
Washington Post was not clear which Ki -Tech forces the Post -Ki ask, but in general it won’t bother you if you are looking for it. Based on the prevailing warnings about its use, it seems more as a place for research than as a direct newspaper.
