One way or the other, we're on the verge of a media revolution thanks to generative AI. Here are specific, detailed proposals for how LLMs can make the media dramatically better.
Jun 15, 2023·edited Jun 16, 2023Liked by Jon Stokes
There was a page made about fixing AI with journalism recently noting the market opportunity for mainstream news to regain the public's trust, along with other ways to aid the news media like making local journalism more efficient and with "pair journalism" pairing AI with journalists akin to "pair programming". Just an excerpt on the market need to fix a broken product:
'A study by Gallup and the Knight Foundation found that in 2020 only 26% of Americans reported a favorable opinion of the news media, and that they were very concerned about the rising level of political bias. In the 1970s around 70% of Americans trusted the news media “a great deal” or a “fair amount”, which dropped to 34% this year, with one study reporting US trust in news media was at the bottom of the 46 countries studied. The U.S. Census Bureau estimated that newspaper publisher’s revenue fell 52% from 2002 to 2020 due to factors like the internet and dissatisfaction with the product.
A journalist explained in a Washington Post column that she stopped reading news, noting that research shows she was not alone in her choice. News media in this country is widely viewed as providing a flawed product in general. Reuters Institute reported that 42% of Americans either sometimes or often actively avoid the news, higher than 30 other countries with media that manage to better attract customers. In most industries poor consumer satisfaction leads companies to improve their products to avoid losing market share. When they do not do so quickly enough, new competitors arise to seize the market opening with better products.
An entrepreneur who was a pioneer of the early commercial internet and is now a venture capitalist, Marc Andreessen, observed, the news industry has not behaved like most rational industries: “This is precisely what the existing media industry is not doing; the product is now virtually indistinguishable by publisher, and most media companies are suffering financially in exactly the way you’d expect..” The news industry collectively has not figured out how to respond to obvious incentives to improve their products. '
I'm all for AI/LLM doing things that allow a journalist the time to do better journalism - ie. better research, deeper stories, etc.
What you describe here is thought-provoking and I can get behind most of it. There are two things that bother me. One is transparency and provenance of the end result. Two is bias.
Dealing with the first one - a consumer of media must be able to establish the provenance of what they are consuming from start to finish. On other words, the who, what, when of the entire process you describe. This is not difficult to do if you're prepared to think a little differently about content and data.
Dealing with the second one, bias, is much more difficult. There is some really interesting research about the consequences of AI generated content being used to feed Llm models. Whatever bias is created intensifies and builds upon itself. Again the integrity and provenance of "data in" and "data out" plays a role here but I also think there are some new human roles that must be created.
Alas, I think you've given short shrift to the fundamental problem that investigative journalism is expensive, but hot-take punditry is cheap. While you note the caveats, sadly almost all that you assert has the same failure-mode we recently saw with blogs, website, etc.
Right here: "... but it would open up the news reporting process to more people and improve the quality of the stories we're able to tell given the current time and budget constraints the media operates under."
Last time around, THIS IDEA FAILED MISERABLY, HORRIBLY, a complete and utter con-job. It was: "Anyone can set up a website, a blog, you-yes-YOU have a printing press! No gatekeepers! There will be a revolution of citizen journalism!"
What happened was that there were a few very small outliers that the hustlers/grifters touted for their self-promotion on the conference circuit, and the industry itself was decimated in the pursuit of clicks.
Look, I love your explanatory stuff, but "we tried it your way, and it didn't work!".
That "next-generation story factory" is going to be one where the culture war and the hot takes are LLM-generated. Because that's the obvious incentives, and that's what the LLM's are extremely good at doing.
This does sound like a big improvement on what we have now, but isn't it a bit like improving the horse and buggy by making a gasoline powered horse to pull the buggy around? If headlines are designed to be misleading and stories are designed to be filler, what we really need is something intermediate in length that shows actual insight in as brief a form as possible.
I’m looking for partners on my project. www.briefer.news
I’m aggregating and summarizing, and looking to create meta summaries by region and subject.
Interested?
Biomant@protonmail.com
There was a page made about fixing AI with journalism recently noting the market opportunity for mainstream news to regain the public's trust, along with other ways to aid the news media like making local journalism more efficient and with "pair journalism" pairing AI with journalists akin to "pair programming". Just an excerpt on the market need to fix a broken product:
https://societyandai.com//p/fix-journalism-using-ai
'A study by Gallup and the Knight Foundation found that in 2020 only 26% of Americans reported a favorable opinion of the news media, and that they were very concerned about the rising level of political bias. In the 1970s around 70% of Americans trusted the news media “a great deal” or a “fair amount”, which dropped to 34% this year, with one study reporting US trust in news media was at the bottom of the 46 countries studied. The U.S. Census Bureau estimated that newspaper publisher’s revenue fell 52% from 2002 to 2020 due to factors like the internet and dissatisfaction with the product.
A journalist explained in a Washington Post column that she stopped reading news, noting that research shows she was not alone in her choice. News media in this country is widely viewed as providing a flawed product in general. Reuters Institute reported that 42% of Americans either sometimes or often actively avoid the news, higher than 30 other countries with media that manage to better attract customers. In most industries poor consumer satisfaction leads companies to improve their products to avoid losing market share. When they do not do so quickly enough, new competitors arise to seize the market opening with better products.
An entrepreneur who was a pioneer of the early commercial internet and is now a venture capitalist, Marc Andreessen, observed, the news industry has not behaved like most rational industries: “This is precisely what the existing media industry is not doing; the product is now virtually indistinguishable by publisher, and most media companies are suffering financially in exactly the way you’d expect..” The news industry collectively has not figured out how to respond to obvious incentives to improve their products. '
I'm all for AI/LLM doing things that allow a journalist the time to do better journalism - ie. better research, deeper stories, etc.
What you describe here is thought-provoking and I can get behind most of it. There are two things that bother me. One is transparency and provenance of the end result. Two is bias.
Dealing with the first one - a consumer of media must be able to establish the provenance of what they are consuming from start to finish. On other words, the who, what, when of the entire process you describe. This is not difficult to do if you're prepared to think a little differently about content and data.
Dealing with the second one, bias, is much more difficult. There is some really interesting research about the consequences of AI generated content being used to feed Llm models. Whatever bias is created intensifies and builds upon itself. Again the integrity and provenance of "data in" and "data out" plays a role here but I also think there are some new human roles that must be created.
This is a great thread .. to be continued?
Alas, I think you've given short shrift to the fundamental problem that investigative journalism is expensive, but hot-take punditry is cheap. While you note the caveats, sadly almost all that you assert has the same failure-mode we recently saw with blogs, website, etc.
Right here: "... but it would open up the news reporting process to more people and improve the quality of the stories we're able to tell given the current time and budget constraints the media operates under."
Last time around, THIS IDEA FAILED MISERABLY, HORRIBLY, a complete and utter con-job. It was: "Anyone can set up a website, a blog, you-yes-YOU have a printing press! No gatekeepers! There will be a revolution of citizen journalism!"
What happened was that there were a few very small outliers that the hustlers/grifters touted for their self-promotion on the conference circuit, and the industry itself was decimated in the pursuit of clicks.
Look, I love your explanatory stuff, but "we tried it your way, and it didn't work!".
That "next-generation story factory" is going to be one where the culture war and the hot takes are LLM-generated. Because that's the obvious incentives, and that's what the LLM's are extremely good at doing.
This does sound like a big improvement on what we have now, but isn't it a bit like improving the horse and buggy by making a gasoline powered horse to pull the buggy around? If headlines are designed to be misleading and stories are designed to be filler, what we really need is something intermediate in length that shows actual insight in as brief a form as possible.