Category: Media

COMPUTERS CAN’T CRY
A computer can tell you how old she is. It can tell you where she lives.
How big her family is. How much money she makes. Or where she shops. What she buys. How much she spends.
And what she reads and when she watches television.
But a copywriter like Greta Nissen can tell you what makes her cry.
Or laugh. Or buy.
Young & Rubicam likes computers.
Young & Rubicam loves Greta Nissen.

Note: I didn’t write the following. I prompted it. In part as an experiment, but also to show how easy it is to be lazy and have AI dunk on itself. I have thoughts on “AI” that go beyond this, certainly, but figured this would be a fun exercise.

I used the following prompt with ChatGPT4o to request an outline, then asked it to write each section. I have made no edits to the output.

You are an amateur writer who’s been blogging for more than twenty years on a variety of topics including current events, technology, music, entertainment, and more. Give me an outline of a piece on how AI writing is terrible, lazy, and going to lead to the internet getting worse when it comes to content and people finding genuine information.

Introduction

In recent years, the advent of artificial intelligence (AI) has revolutionized various industries, with writing and content creation being no exception. AI writing technology, powered by sophisticated algorithms and machine learning, can now generate articles, blog posts, and even creative stories at an unprecedented speed. This rapid adoption of AI in writing has been driven by its promise of efficiency and cost-effectiveness, making it an attractive option for businesses and content creators alike.

However, as AI-generated content becomes increasingly prevalent, it is essential to scrutinize its impact on the quality and authenticity of information available on the internet. While AI writing tools offer convenience, they also pose significant challenges that could undermine the very essence of meaningful and trustworthy content. The reliance on AI to produce written material is leading to a decline in content quality, originality, and the human touch that makes writing truly engaging and informative. In this blog post, we will explore the dark side of AI writing, examining how it contributes to a flood of mediocre content, spreads misinformation, and ultimately threatens the integrity of the digital landscape.

The Problem with AI-Generated Content

Lack of Originality and Creativity

AI generates content by analyzing vast amounts of existing data, identifying patterns, and constructing text based on those patterns. While this method allows AI to produce coherent and grammatically correct writing, it inherently lacks originality and creativity. AI does not generate new ideas or offer fresh perspectives; instead, it recycles information and rephrases it in a way that can often feel repetitive and formulaic.

For instance, many AI-generated articles follow a predictable structure and use common phrases that make the content feel generic. This lack of creativity is evident in the way AI handles complex topics, often resorting to surface-level explanations without the depth and insight that human writers bring to their work. Human writers, driven by their unique experiences, knowledge, and creative instincts, craft content that resonates with readers on a deeper level, offering original ideas and engaging narratives that AI simply cannot replicate.

Quality and Accuracy Issues

One of the significant limitations of AI in writing is its inability to fully understand context and nuance. While AI can process large amounts of data, it struggles to grasp the subtleties of language, cultural references, and the intricate details that often make writing accurate and compelling. This limitation leads to a higher likelihood of factual errors and misleading information in AI-generated content.

For example, an AI might misinterpret a historical event or misquote a source because it lacks the human capability to critically analyze and verify information. Such inaccuracies can be particularly problematic for readers who rely on online content for reliable and precise information. The spread of misinformation, whether intentional or not, can erode trust in digital content and make it challenging for readers to discern fact from fiction.

Flood of Mediocre Content

The ease and speed with which AI can generate content have led to a significant increase in the volume of online material. However, this surge in quantity has not been matched by a corresponding increase in quality. The internet is becoming saturated with mediocre, cookie-cutter articles that offer little value to readers.

As a result, discerning valuable content amidst the noise has become increasingly difficult. High-quality work and important voices risk being overshadowed by the sheer volume of AI-generated material. This dilution of content quality poses a significant challenge for readers seeking insightful, well-researched, and engaging articles.

Loss of Human Touch and Personal Voice

The essence of compelling writing lies in the personal experiences, unique perspectives, and genuine emotions that human authors bring to their work. AI, devoid of consciousness and personal experiences, cannot replicate this human touch. The result is content that feels sterile and impersonal, lacking the warmth and relatability that make human stories so powerful.

For instance, consider a memoir or a personal blog post where the author’s voice and experiences are central to the narrative. AI cannot authentically convey such stories or the emotions behind them. It can only simulate a superficial version, devoid of the depth and nuance that come from genuine human experiences. This absence of a personal voice and emotional connection diminishes the overall impact and authenticity of the content, making it less engaging for readers.

Consequences for the Internet and Information Consumption

Erosion of Trust in Online Content

As AI-generated content proliferates, readers are becoming increasingly skeptical about the authenticity of online articles. The realization that a significant portion of content is produced by machines rather than humans can lead to doubts about the credibility and trustworthiness of the information presented. This skepticism is further compounded by instances where AI-generated content has contributed to misinformation.

For example, AI systems, when fed biased or incorrect data, can produce articles that perpetuate falsehoods or present skewed perspectives. A notable case involved an AI-generated news article that misinterpreted a complex legal issue, leading to widespread confusion and the dissemination of incorrect information. Such incidents undermine the reliability of online content and can have serious repercussions for public discourse and knowledge.

Challenges for Content Creators

The rise of AI-generated content poses significant challenges for human writers and journalists. One of the most pressing issues is the increased competition from AI, which can produce large volumes of content quickly and at a lower cost. This competition can make it harder for human writers to secure work and maintain their livelihoods, potentially leading to job losses in the writing and journalism industries.

Moreover, the devaluation of human-authored content is another critical concern. As AI-generated articles flood the internet, the unique value of content crafted by human hands may be overlooked or undervalued. Readers and platforms might prioritize quantity over quality, favoring the endless stream of AI-generated material over thoughtful, well-researched pieces by human authors. This shift can diminish the appreciation for the skill, creativity, and effort that go into producing high-quality human-authored content.

Impact on Search Engines and Content Discovery

AI-generated content also poses a threat to the efficacy of search engines and the process of content discovery. Search engines rely on algorithms to index and rank content, but these algorithms can be manipulated by the sheer volume of AI-produced articles. As a result, search results may become dominated by low-quality, keyword-stuffed content that lacks substance and value.

For instance, an AI can generate hundreds of articles on a trending topic, each designed to rank highly in search engine results through strategic use of keywords. This practice can push genuinely informative and well-researched human-authored articles further down the search results, making it harder for readers to find valuable content. Consequently, the overall user experience on the internet is diminished, and the quality of information available to the public is compromised.

In summary, the widespread adoption of AI writing technology has far-reaching consequences for the internet and how we consume information. The erosion of trust in online content, challenges faced by human content creators, and the impact on search engines and content discovery all point to a troubling trend. If left unchecked, these issues could lead to a digital landscape where quality and authenticity are sacrificed for the sake of convenience and efficiency.

Possible Solutions and a Call to Action

Promoting Ethical Use of AI in Writing

To mitigate the negative impact of AI-generated content, it is crucial to establish guidelines for the responsible use of AI in content creation. These guidelines should emphasize the importance of quality over quantity, ensuring that AI tools are used to enhance, rather than replace, human creativity and insight. Additionally, transparency is key: platforms and publishers should clearly disclose when content is AI-generated. This transparency allows readers to make informed decisions about the credibility and authenticity of the material they consume.

Supporting Human Writers

In the face of increasing AI-generated content, it is essential to support human writers and prioritize their work. Platforms and publishers can play a significant role by valuing and promoting human-authored content. This support can take the form of financial incentives, featuring human writers prominently, and creating initiatives that help writers adapt to the changing landscape. Encouraging collaboration between AI and human writers can also lead to innovative and high-quality content that leverages the strengths of both.

Improving AI Technology

Ongoing research and development are vital to address the current limitations of AI in understanding context, nuance, and the human touch. AI developers should work closely with content creators to enhance the quality and reliability of AI-generated content. By focusing on improving AI’s ability to comprehend and replicate human creativity and authenticity, we can ensure that AI serves as a valuable tool for writers rather than a replacement.

Conclusion

In this blog post, we have explored the dark side of AI writing, highlighting its negative impact on content quality, originality, and authenticity. AI-generated content, while efficient, often lacks the creativity and personal touch that human writers bring to their work. It can lead to misinformation, an overwhelming flood of mediocre content, and a loss of trust in online information. Moreover, the rise of AI writing poses significant challenges for human content creators, from increased competition to the devaluation of their work.

To counter these issues, we must promote the ethical use of AI in writing, support human writers, and continuously improve AI technology. By establishing guidelines for responsible AI use, prioritizing human-authored content, and enhancing AI’s capabilities, we can preserve the quality and authenticity of online content.

Ultimately, the future of digital content depends on the collaborative efforts of readers, writers, and tech developers. Together, we can ensure that the internet remains a space where quality, creativity, and the human touch are valued and preserved. Let us take action now to build a better digital future where technology complements, rather than compromises, the integrity of our online information.

Over at NiemanLab, Laura Hazard Owen covers “the long, complicated, and extremely frustrating history of Medium” and I’m still trying to figure out what problem Medium was trying to solve. I guess acting as a hub for folks who want to write longform pieces but not maintain their own websites?

“They believed if you write it, they will come,” one employee told me. “And it never worked that way.”

Via The Mess At Medium

The pay model was (is?) interesting, but with Substack* and other services giving authors a way to monetize and distribute their work the model may need to shift. Yesterday Medium announced an overhaul to their editorial strategy, so maybe the shift has begun?

Or not:

Meanwhile, the company will continue to rely on Google and Facebook traffic to generate hits it can convert into paid subscribers. The acquisition of the e-book publisher, Glose, is intended to create a bigger library of “evergreen” content on Medium that will drive more traffic to the site via search engines. Like Blogger and Twitter before it, Medium will bet on unpaid labor and algorithms.

Check out the whole piece at The Verge to get a deeper dive into the voices behind Medium’s success and where it stands now.

As someone who enjoys longform writing, both as a consumer and not-as-often-as-I’d-like producer, I’ve liked the idea of Medium but I’m not sure if it is a genuine need being filled or trying to convince folks “hey, longform can be cool too!” in a world of tweets and headline skimming.

Throw into that the fact that its model seems to be dependent upon self promoted user generated content feeding third-party algorithms and you really just have a new way of doing Buzzfeed or The Huffington Post from ten years ago. All of which runs contrary to the initial premise of not to sweat SEO and metrics and analytics because the cream will rise to the top and that rising tide will lift all boats in the Medium pond.

The idea of elevating, feeding, promoting, and paying good content and its creators is a noble venture. Trying to figure out a way to do that successfully – both in informing but also profitability – is the Internet nut so many have been trying to crack for more than twenty years now. Too often the former, informing, is thrown by the wayside for the latter, profitability (see Buzzfeed and The Huffington Post model ten years ago). And you can try to go back to the former once you succeed at the latter, but then credibility (or lack thereof) comes into play.

Maybe, in the end, the unspoken harsh truth is people just don’t care anymore.

Real writing, journalism, thoughtful longform written content is now a niche instead of the norm. Perhaps the secret is just figuring out how to help individuals find their 1,000 true fans and letting the world rebuild from there.

Or maybe Medium will figure it out?

This Medium rabbit hole was inspired by Guy Gonzalez, who you should read more often.

* I have an only half-joking running debate with Kenney the Elder that Substack is essentially OnlyFans for writers in that it’s a platform for folks to monetize what social media encourages folks to give away. Perhaps that’s a post for another day…

11 years ago Haaretz experimented with replacing journalists with authors.

Even as it shrinks, the national media is reorganizing around a social media–to–cable news pipeline of daily outrage. It is shedding the skin of its once-sacred “view from nowhere” objectivity and embracing the benefits of cruder ideologies. It wants eyeballs, but it doesn’t want to pay for material. Why do that when a generation of strivers will do it for free, or close to it? No, it’s not so hard not to see Andy Ngo as one vision of the journalist of the future, self-employed in an Uberized model that gobbles up inflammatory content and takes no responsibility for how it’s gathered. These media workers will be ambitious, ideological, incurious, self-promoting, social media native, willing to force the story, and very, very vulnerable.

Joseph Bernstein “Andy Ngo Has The Newest New Media Career. It’s Made Him A Victim And A Star.”

“Why wasn’t I consulted,” which I abbreviate as WWIC, is the fundamental question of the web. It is the rule from which other rules are derived. Humans have a fundamental need to be consulted, engaged, to exercise their knowledge (and thus power), and no other medium that came before has been able to tap into that as effectively.

Paul Ford – “The Web Is a Customer Service Medium”

Paul Ford’s entire piece from 2011 is an interesting observation on the web as a medium and how newspapers and others have failed to truly utilize it. I’m going to have to file away Gutenbourgeois for later use.

Turn your readers into members. Not visitors, not subscribers; you want members. And then don’t just consult them, but give them tools to consult amongst themselves. These things are cheap and easy now if you hire one or two smart people instead of a large consultancy. Define what the boundaries are in your community and punish transgressors without fear of losing a sale. Then, if your product is good, you’ll sell things.

He’s essentially talking about the democratization of news and information, citing Wikipedia and MetaFilter as examples of the wisdom of the crowds, but with filters and guide rails in place that reward good and limit bad to an extent.

In each of these, the audience participates in what is essentially the final product and ultimately determines its value.

The challenge for the press and the internet in general is to make sure there are those “one or two smart people” who can help manage these communities. Otherwise you get the firehose that is Facebook and we’ve already established what a poor job Facebook does as gatekeeper of information.

A larger conversation could be had about this piece and how well it holds up in time in an era of Trump, Russian meddling, “Fake News,” and other real world consequences of how information is spread and believed. I’ll admit that in 2009 I was dismissive of Andrew Keen’s sounding the alarm on the consequences of Virtual Democracy:

Keen displays a complete lack of faith not only in individuals to make right and rational decisions, but also in communities to do the same.  But this also smacks of a fear from those in power in empowering those without power, an argument that has been made against Keen for a while as he has argued that there is a need for “gatekeepers” for information.  When gatekeepers merely fact check this is not a problem. But if they seek to limit access to information, you deny people the right to be fully informed (or even mis-informed, such is life).

How quaint.

But what does stand the test of time from Paul Ford’s piece is the view that the fundamental question of the web is “Why Wasn’t I Consulted?” (WWIC).

Look around. Entire websites are devoted to people who feel their voice should be heard. Twitter exits because people have opinions and snark and memes to share that they feel should be entered as legitimate points within a larger debate.

Heck, I’ve been blogging on and off for nearly 18 years because no one asked for my opinion on these things.

There’s that saying about what opinions are like and everyone having one after all.

WWIC as a question is the foundation of townhall meetings that invite public discourse on community topics. Only now the Internet exists as a virtual townhall for the entire world on every topic big and small.

The good and bad of the Internet is that it presents the ability for all opinions to have a chance at equal platforms, while in reality not all opinions are necessarily equal. There is no technical difference between a tweet from an account created a week ago with two followers than one from, say, Barack Obama. Technically speaking, one has the same opportunity to germinate and blossom and be found and read as the other. And sometimes, that random tweet from an account that still has an egg as a profile picture, will blow up into something significant, even if it has no foundation in truth, and find its way into the public consciousness far quicker and deeper than the truth might later on.

Lacking some sort of editorial oversight, those “one or two smart people” or even the Gutenbourgeois, there’s little to stop bad ideas from changing our understanding of something in ways that is no longer true, but now becomes “reality” because the wisdom of the crowd wants it to be.

Keeping WWIC in mind and even appending it to a viewpoint you find online is useful because it can help put some of these random voices into a context. Is that voice shouting out because it comes from a place of informed opinion that has true value? Or is it simply looking for attention like an angry uncle on the front lawn yelling at the sky?

The caution I’d advise is against using WWIC to dismiss what is an otherwise valid yet opposing viewpoint. Just because someone wasn’t consulted doesn’t mean they can’t add value to a topic. And we should want to embrace opposing viewpoints to have a more well rounded understanding of the world. It’s just becoming more important that we strive to become our own fact checkers, because in a world of democratized information there’s no one to check it for is.

Somewhat related sidenote:Remember the fanfare when Facebook was partnering with the Associate Press and Snopes to fact check stories on the social network and help combat fake news? Yeah, those guys are out.

What were the algorithmic criteria that generated the lists of articles for us to check? We never knew, and no one ever told us.

There was a pattern to these repeat stories though: they were almost all “junk” news, not the highly corrosive stuff that should have taken priority. We’d be asked to check if a story about a woman who was arrested for leaving her children in the car for hours while she ate at a buffet was true; meanwhile a flood of anti-semitic false George Soros stories never showed up on the list. I could never figure it out why, but perhaps it was a feature, not a bug.

Brooke Binkowski – “I Was A Facebook Fact-Checker. It Was Like Playing A Doomed Game Of Whack-A-Mole.”

Over at Karigee is an interesting highlight from a Q&A of Sarah Anne Johnson interviewing Amy Bloom in Conversations with American Women Writers (how’s that for old school blogging link credit!):

Q: You often use humor to diffuse an emotionally intense situation, and at the same time to create pathos, or a sense of the real sadness underlying the attempt at humor and the need for humor in a given situation. For instance, in “Silver Water,” the scenes with the therapists, especially Big Nut, are funny in spite of the gravity of the situation. Are you aware of this as you work, or does it just come out at certain times? How does humor work in your fiction?

A: I don’t see that much as diffusing the sadness of the situation. There is humor in grief. Funny things happen in hospitals. That’s just how it is. I don’t think that life is composed of sad moments, which are sad, in which bad things have happened to good people, and happy moments, in which good things have happened to good people. So for me, there being humor in the midst of difficulty and pain is not an attempt to either lighten the pain, or change the focus, or make a comment on it. It’s the way it is. To me it’s no different than the idea that there are both flowers and weeds in the garden. I don’t feel like if I see weeds in the garden, I think, That’s an interesting comment on the flowers. I think, That’d be because it’s a garden.

I appreciate this take from both a storytelling and real world perspective.

Humor in grief or when facing a serious situation isn’t uncommon or out of place – it’s how people cope, it’s how people face what may normally be absurd, but it’s also just how the world works. Emotions don’t live in a silo. We don’t live in a silo.

Somewhat related (apologies to Amy Bloom if she’s offended I’m about to compare her statement on her literary work to superhero films) is a debate among people who need to lighten up about the major differences or flaws between the Marvel and DC Cinematic Universes. Marvel uses humor to drive the plot, develop a character, or just to lighten up a scene. DC, in the other hand, fills in the gaps with grunts and punches.

There are some who argue that Marvel’s approach is unrealistic and that may be true to its degree, but the DC approach isn’t how human nature works either. As serious as the world is around us, there’s something absurd, someone to say something out of place, someone to ease the tension.

When I see people argue against any humor in a tense situation, I’m reminded of a bit in Mark Bowden’s Black Hawk Down which chronicles the 1993 Battle of Mogadishu after the downing of two US Black Hawk helicopters.

Fire was coming from all directions, but mostly straight up and down the alley. They were still expecting the arrival of the ground convoy at any moment. They had no way of knowing that the convoy was lost and taking heavy casualties.

Fales was too busy shooting from his position out by the tail to take notice of the placement of the floor panels. He had a pressure dressing on his calf and an IV tube in his arm.

“Scott, why don’t you get behind the Kevlar [floor panels]?” Wilkinson asked. Fales looked startled. Only now did he notice the barricade.

“Good idea,” he said.

Crouched down behind the panels, Wilkinson and Fales watched as the intense gunfire ripped first one hole through the tail boom, then another. Then another.

Wilkinson was reminded of the Steve Martin movie The Jerk, where Martin’s moronic character, unaware that villains are shooting at him, watches with surprise as bullet holes begin popping open a row of oil cans. Wilkinson shouted Martin’s line from the movie.

“They hate the cans! Stay away from the cans!”

Both men laughed.

The world going to hell around them and they laughed. It doesn’t belittle or negate the chaos they’re in. It’s just the way it is.