Imagine this: We live in an age of no truth. Millions of media platforms feed us personalized information the moment it uploads without regard to its validity. An explosion in the nation’s capital; masked bandits dropping off boxes of ballots; the new president hugging a war criminal—is any of it real? Images used as “proof” have become a thing of the past.
Some may go out of their way to seek the truth and ensure the information has a verified source—a witness or first-hand account—before believing it. The vast majority of published information, however, is the result of artificial intelligence (AI) scraping those original ideas and reassembling them into millions of different formats for billions of viewers. Those willing to sift through the 99% of AI-produced information to do that will be few.
Unable to compete with the speed and abundance of auto-generated content, publishers won’t be able to afford the costs of verifying information. Will there even be a demand for the truth? If they share misinformation, they lose credibility. Without anyone questioning it, whatever the AI platform says becomes fact over fiction. Nothing we consume can be trusted.
Sound scary? Without restraints on AI-generated content, this picture may soon seem less like science fiction and more like reality.
Those working closely with AI development are already sounding the alarm. In March of this year, over 1,000 researchers and tech leaders, including Elon Musk and Steve Wozniak, called to halt the development of AI systems beyond ChatGPT-4 until “we are confident that their effects will be positive and their risks manageable.” Even more recently, Google’s “godfather of AI” stepped down and warned the public about ethical fears.
OpenAI researchers estimate that AI could disrupt at least 10% of work tasks for 80% of the U.S. workforce, and this process is already affecting publishing. The Associated Press already uses AI to autogenerate corporate earnings reports, recap sporting events, and transcribe live events. In China, news agencies deployed AI news anchors capable of engaging in propaganda-driven conversations. UK media outlets can subscribe to the AI-powered agency, RADAR, and publish any of the thousands of local stories it generates a week.
Information everywhere, but none that we can trust: To avoid this dystopian future, we need to do more to protect the integrity of published information.
WILL DIGITAL PUBLISHING BECOME OBSOLETE?
AI is nothing new. For years, machine-learning algorithms have powered voice assistants, search algorithms, and facial recognition software to improve our work and ease of life. But with the entrance of generative models, AI went from a vision on the horizon to a reality upon us. Chatbots can now hold conversations, write computer code, and even pass the bar exam. And this is only the beginning.
Will there be a place for publishers in this new world?
AI tools like ChatGPT crawl websites for content used to respond to a user’s inquiry. Essentially, publishers invest time and contribution to online content for generative models to steal and capitalize on without providing credit (or revenue) to the publisher. Webpages may become obsolete as users search for information by going directly to this new generation of AI chatbots like Bard, ChatGPT, and Bing Chat.
Social platforms have already negatively impacted publisher revenue. The question is: Will AI follow suit? We can expect many users to visit generative bots for information rather than referring to publisher sites directly, which in turn results in a substantial loss of revenue for publishers.
Some of the dangers of unregulated AI in publishing derive from the lack of scarcity.
In a world without scarcity, it becomes much harder to stand out or verify the legitimacy of information. When content can be created effortlessly, we can expect thousands of new publishers to crop up and the value of content will diminish. These new publishers will fight for the same amount of advertising budgets that we have today and every individual publisher will inherently earn less.
A HUGE OPPORTUNITY FOR PUBLISHERS
AI is our new reality, and it can already make a journalist’s job easier. Today, AI can perform administrative tasks, reduce research time, help with translations, act similarly to an intelligent intern or personal assistant, and more. Improved efficiency will translate into powerful differences, such as smarter investigative work, enhanced creativity, better research, and overall stronger output. This is especially true for local media where newsrooms lack resources. Content creators can focus on better defining their voice, inventing unique story angles, and honing in on their distinctive style.
To keep the flow of quality information as accessible and discernable as possible, we need:
1. Transparency From All Ends
Everyone needs to agree to expose audiences to what is true and what is not. Publishers should insist on regulations that establish powers of authentication over creative input and disclose the use of ChatGPT in their content. Many publishers, including The Guardian and Insider, are already starting to establish guidelines around AI. Advertisers should aim to be more aware of where they advertise and avoid sites that publish fiction as fact.
2. A Value Exchange For Published Content
If a media outlet publishes a video of an NBA game that aired the night before without permission or credit, they can expect a letter citing copyright infringement. Original written work needs protection and guardrails. AI companies scraping original work without providing any compensation to creators is not sustainable.
3. More Knowledge
Publishers should be very proactive in adopting and experimenting with AI technologies. They can stay ahead of the game and accumulate knowledge to influence developments and policies around AI that may impact them.
With AI growing in popularity across the internet, publishers should be a part of the conversation. The tech engines behind generative bots must put publishers and their content at the forefront and think about the impact they are making on the industry they so desperately need to survive.
Tom is the Co-Founder and CEO of EX.CO.