Subscribe to discover Om’s fresh perspectives on the present and future.
Musings on Media in the Age of AI
I recently wrote about the future of the browser and Surf, a new app from the creators of Flipboard. Both stories explore the changing nature of the web and its impact on the media landscape. I’m not shy about expressing my frustrations with the establishment media and the ever-present gulf between technology and old media companies. I’ve been involved with the internet and online publishing from the very beginning — even before many big media companies embraced it. Those companies have almost always lagged behind in understanding the shifting reality of what media is, how we consume it, and how it’s distributed.
The widespread lack of understanding among the establishment media allowed Google to become a behemoth. And then they essentially handed the keys to the kingdom to Mark Zuckerberg and Co. They are making the same mistake with the new AI platforms. Despite all the brouhaha about fighting Perplexity or OpenAI, it will amount to nothing more than a short-term squeeze play.
Some media companies are simply signing deals, taking the cash while they can. By doing deals with these platforms and giving them access to their content, most publications might feel they will be rewarded with traffic to their websites. Here’s what Nicholas Thompson, CEO of The Atlantic, said on a podcast:
We can go through it in complex way or the simple way. The simple way is we believe it provides revenue, but more importantly provides a potential traffic source. Provides an avenue for a product partnership that could be very beneficial, and that provides a way for us to help shape the future of AI…
…[Another] part is this very interesting search element, where right now in OpenAI they have browse mode and they can link out to Atlantic stories. They have said that they’re going to build a search product. They have not launched the search product, but they have said they would build it. We have allowed them to include The Atlantic in their search product.
Our view is that if this becomes an important way that people navigate the internet, that it will be better for us to be in it than to not be in it, and also to help shape it than not help shape it.
The Atlantic and its brethren have about as much chance of shaping the future as I do of convincing Elon Musk to focus on playing cricket. The challenge for these publications is more existential.
Unlike search engines, AI platforms are built on precision and summaries. They’re unlikely to be a major source of traffic or advertising revenue. What about getting paid for the summaries OpenAI serves in response to prompts? These prompts will vary widely, as will the responses, breaking the traditional mass-media revenue model. As I explained in my “future of browser” article, information itself is being atomized, which will likely upend the web and media as we know them.
Facebook made big promises to media companies, only to pull the rug from under them. OpenAI, Anthropic, and Perplexity are likely to be equally, if not more, ruthless. These platforms won’t need the media for long.
The old media has consistently misunderstood digital transformation, and it’s no surprise that we have a media ecosystem still trapped in old monetization models, where “interruptions” have only grown more aggressive. What began as occasional magazine ads has evolved into a constant barrage across all platforms—from billboard-cluttered webpages to podcast sponsorship breaks and algorithmic social feeds designed for ad delivery.
Social media platforms, built around algorithmic feeds and advertising models, have reduced content discovery to a game of clicks, likes, and engagements. Mass-market media has followed suit, optimizing for sensationalism rather than depth. All of it, from podcasts to news apps, interrupts users constantly with ads, pushing all of us to exhaustion.
In reality, the seeds of media’s destruction are built into its architecture, because outlets must feed advertising systems, not the audience. The media establishment disregards why audiences visit them, and it’s no surprise the system has reached its limits. Too many advertisements, too many interruptions, and too much “content” mean that, as an end customer, you are decoupled from media brands.
Nowhere was this more obvious than during the recent election cycle. The news cycle showed that information consumers were ready to move on from tired old content creation, delivery, and distribution models. Who wants to deal with 500 versions of a 500-word article with a sizzling headline and a sliver of actual news? Try sitting through 10 minutes of CNN or any cable show. Screaming, shouting, and superficiality make it easy to tune out traditional media.
It’s hardly surprising people want 60-second summaries on TikTok and headlines on Twitter. A recent Pew Research study found that “about one-in-five Americans – including a much higher share of adults under 30 (37%) – say they regularly get news from influencers on social media.” While social media is known for hot takes and memes, it’s much more than that, as Pew’s study revealed. “Overwhelming majorities say they get all four types of content asked about in the survey: basic facts (90%), opinions (87%), funny posts (87%), and breaking news (83%),” the survey noted.
OpenAI and Claude are demonstrating that we’re entering a phase where individuals will engage with platforms as a singular individual entity. Even if these platforms adopt advertising, each query will generate a unique response, making it difficult to replicate traditional mass media monetization. This represents the newest variation of the game played by Google and Facebook.
Google and Facebook are already showing that by embedding “Ask AI” and “Summaries” into their products, they are ready to produce one-to-one advertising using “Generative AI.” They won’t need to send people anywhere, much like OpenAI or whoever triumphs as a big winner in the AI sweepstakes.
The internet was originally envisioned as a place for connection, collaboration, and discovery. But over time, it has been distorted by business models that prioritize engagement metrics over meaningful interaction. Discovery has long been the open web’s greatest challenge, with search engines turning it into an SEO game and social platforms creating algorithmic echo chambers. AI platforms are making discovery almost irrelevant.
You stay still, but your AI agent goes out and fetches, distills, and synthesizes the content and renders it in whatever format you want — audio, video, or text. This is the future. None of the media business models will work in the future — neither advertising nor paywalls. Today’s content deals, like the one The Atlantic signed with OpenAI, are akin to the sugar high you get from soda. The sugar high is followed by the inevitable crash.
December 21, 2024 San Francisco
Addendum: The Washington Post is drama central these days. The paper is trying to come up with a new growth strategy. And there is a new (editor in) chief in town. Furthermore, some believe they can save the paper if they can buy it from Jeff Bezos. I wish all of them good luck in all those endeavors. After all, trying is better than doing nothing. The media industry is in a dire situation. It is time to reinvent. and it starts by rethinking the meaning of media in the age of “AI” and then we need to figure out how to make it work financially.
15 thoughts on this post
Comments are closed.
Over time, I have massaged my information intake. Presently…
Newspapers: NYTimes, Atlanta Journal and Constitution (local)
Magazines: New Yorker, The Atlantic
Radio/TV: NPR, PBS
Email: Robert Reich, Judd Legum; Ed Zitron; Fix the News; NextDraft; The Athletic.
I do not use, nor want, my information summarized.
I do skim and may cut short contents. I do not use audio because it takes time to listen. All content should be dismissable to save time.
Those are good sources of information, though I much prefer to read a wider variety of “sources.” The fact remains that the future of this industry is going to be determined by those who are growing up on TikTok and those who will use “AI” natively. That will definitely up-end the industry as we know it. A lot of our media infrastructure is built on “traffic” thinking and not qualitative thinking.
In keeping with your “dark thinking”: Advertising has broken the dream of the internet .
Is the internet now is really more of the trope “if it bleeds it leads”? Most of us do not know what news is, neither do the providers of news. How big of a stunt do you need to do for clicks? How fast do you have to be to stand out in front of the noise, How accurate?
The internet is an information delivery tool. IP networking has no content filter for truth or lies it just delivers packets. Your “social network” is a concocted morass of someone’s dream of making money for them or their company. Many newsrooms are now just targeted marketing arms of their parent corporations. Or agenda delivery tools for the .05%.
So I take a short journey every day to find reality on the Internet. Most days it is not there.
I am trying hard to find where I can get reporting that makes some sense. Selecting sources for content has been hard. I have found two of interest. I pay these folks directly for content: The Verge and 404Media.
I read a few RSS feeds that fit my technology, information security and photography news. A little Daring Fireball, DPreview, MacStories, OM and friends, Eric Topal. Sometimes more sometimes less. RSS feeds only here no website surfing, doom scrolling.
I study building science with a targeted Patreon group. I use very little of YouTube to study things I am interested in. Or enjoy the travels of Itchy Boots on her motorcycle travels through the world.
Sometimes all we need is Silence. Sometimes we need more music in our lives. Sometimes we need the touch of a family member.
Bob
First, 404Media is fabulous. I am a big believer in their mission. I love your other selections. Maybe I should start publishing my own list of what to read again? I find there’s good stuff, just not enough and it can easily be surfaced.
As to your meta point, I think Internet has become a mirror image of limitations of “media orthodoxy.” Whether it is 500-churnicles or streaming old TV tropes, it has become victim to their lack of imagination. Hell, even folks like Apple can’t seem to imagine what are the possibilities of their platforms.
If you look at where we are going, you are going to see a very different future. It is not going to be what we have been used to so far.
PS: If you want to enjoy some good quality YT videos, follow Morten Hillmer (wildlife photographer) and Lost Lakes, an adventurist.
@Om,
Would love to see a post about what/where you are reading. Your “dots” so to speak.
Cheers!
Steve
I am thinking about writing this over the holiday break. It would be a perfect way to end the year.
The Internet is the media, for better or worse.
TV, magazines, newspapers, cable — they’re dead. Or dying. Irrelevant. They’re buggy whip makers in 1924.
Without business models, there’s no business. The question remains the same as I asked while at Medill in 1977. How do we create profitable online business models for ongoing fact-based content?
Dana,
100 percent agree. The internet has turned into the “old media” and now is time to think of new ways of doing media. In a way TikTok is a perfect example of new “media” for new screens, new demographic and new network. Same as what we did with blogging, and later with social.
A few days ago, I saw a short video on China’s WhatsApp aka WeChat: “Famous American journalist angrily criticizes Musk: Stop talking nonsense, what is X platform even calling itself a media?”
The person in the video is Jim VandeHei from Politico → Axios. I can’t think of anyone more suited to say these words.
First, his experience gives him the authority to say it. Second, his arrogant pride feels completely out of place, just like the literal meaning of what he said.
These elites will hold onto their journalistic ideals and take them to the grave, and then we’ll return to a world without news, like the thousands of years human lived before mass media.
Jim is a very smart guy and understand the media better than most people. In a way, Axios is a “traditional” media version of newsletter/social media boom. They use those wrappers to do traditional media, but more updated to the current consumption behavior. However, the thing is that behavior is changing and changing fast. There are limitations on how fast media will adopt and adapt.
The journalistic ideals are great thing to hold on to — how they apply to changing eras and changing economic realities have to be different. I don’t think we return to a world without news. We need to have news — just like we always have. What will change is the “economics” and that means we need to take chances with the new economic model.
I’m too old a reactionary to add anything cogent to this discussion. Old media models depended on the scarcity/high cost of production and distribution as a barrier to competition. The cash cow transitions from 1995 to 2005 was too little, too late, which combined with the hubris of journalism standards, all but ensured the media titans were doomed in the face of the big platforms controlling the time and attention of the audience.
If modern youth are satisfied with TikTok or Twitter or YouTube as their primary news source, then so be it. As for AI going forth and collecting “content” for me — I’m skeptical. Sounds like just another version of smart agents and customized portals serving up a suspicious stew of click bait.
The appeal of media to me lies in accidental discovery. Case in point — last Sunday’s print edition of the New York Times carried a story about a group of dissident nuns in Spain who were squatting in their convent after defying the Vatican. Marvelous story, beautifully told, but not one to be found in my digital streams. Publishing on glass kills the power of page design and typography to emphasize or deprecate stories. I don’t want to fall into a algorithm defined bubble.
As for business models. Publishers who strike deals with the AI vacuum cleaners are the Neville Chamberlains of journalism. I hope the courts slam the door shut on that blatant theft in the new year.
Brilliant example. I found this story similar to many columns in SCMP that I read a lot, attractive and may broaden my world view.
My takes:
1. Good content needs to be discovered and reach the right audience.
2. Good content needs to be rewarded so good creators may keep producing.
3. However, no individual / organization consistently produces good content.
Because:
1. The reputation and expertise of the NYT make us believe it has a higher chance of giving us good stories.
2. We (mostly) don’t worry about the authors not getting paid, because it’s the NYT.
3. Yet, any media subscription may too expensive for us, or bundling subpar contents, which was common in the magazine era.
Inference:
1. Algorithms (developers) once promised to connect content with the right audience but failed us.
2. Whether joining institutions or starting their own Substack, good creators need sufficient earnings to stay motivated.
3. Unfortunately, I’ve seen some journalists turn to independent ventures, only to deliver newsletters that not-so-well or diluted. If I were running a “one-person company” I’d probably have to do the same.
Finally – if Andy Warhol was right – most people might create 1-2 great works, like “The Spanish Nun”, in their lifetime – perhaps telling their own story, or not. Ideally, we’d want everyone to be adequately rewarded for such a “lifetime masterpiece”, maybe even a movie copyright purchase, and then live their life.
The selection mechanism for this could be mail it to the NYT, or Yahoo!, Digg, Facebook, or TikTok (which I feel is closest to this ideal but still far from perfect), or none of the above.
Hi, Om,
I wonder your take on the “positive” sides of AI for media, such as Schibsted saying they use it for first drafts of more rote stories and also to help with some data-driven investigative work.
Dorian,
If I was to predict, we will have an AI enabled media room in five years. It will be moot point by them, though it would also mean a much smaller media industry.
God this is depressing. I find the prospect of AI synthesising my already aggregated content deeply alarming. The internet bought self publication and loss of editorial control. We need provable editorial standards back quickly. And it will happen.