As developments in artificial intelligence progress, this technology is increasingly critical in news publishing. AI is changing the way news publishers curate and generate articles and how the news is distributed to the masses. It’s had a profound impact on news publishing in general, with both positive and negative consequences.
This guide will discuss the role of AI in news publishing and what it means now and in the future. We’ll cover the key issues involved in artificial intelligence and what news publishers can do to embrace this tech positively.
Copyright Infringement Risks and ‘Hallucinations’
Research from Statista in 2023 identified two key stats about AI in the newsroom:
- 22% of news publishers see using AI for news gathering as ‘very important’ (a further 49% see it as ‘somewhat important’)
- 28% of news publishers see using AI for content creation as ‘very important’ (a further 39% see it as ‘somewhat important’)
Both data points indicate a growing willingness and need for AI, to find and produce news stories. While many reporters suggest this improves a newsroom’s productivity, it throws up two major concerns: copyright infringements and ‘hallucinations.’
What are the copyright infringement risks for AI in news publishing?
Concerns exist primarily about AI creating news stories while infringing on the intellectual property rights of other news publishers or creators. Artificial intelligence pulls its stories and ideas from somewhere, and there’s often nothing stopping it from fragrantly ripping off existing publishing. This creates an even bigger problem, as the laws surrounding AI copyrighting are slightly blurry.
As a result, the Copyright Office in the US is launching a new initiative to overhaul copyright law and policy issues raised by AI. It hopes to develop new laws to make this particular issue easier to deal with and should prevent copyrighted AI news pieces from being published.
Over in the EU, there’s the proposed Artificial Intelligence Act, which also looks at how AI is used to create and distribute content. The key finding here is that creators of AI models will no longer be allowed to train these models using copyrighted material found online. It could breach the copyright law, which completely alters how AI is used in newsrooms as it limits the data these models can pull from.
What are ‘hallucinations’?
Hallucinations in AI refer to fabricated news stories that appear real but aren’t. They may be missing key facts or could be fabricated. This is arguably the biggest problem with AI in news publishing, as it can spread fake news like wildfire. NewsGuard reports that the number of new sites with AI-generated false stories has increased 1000% in the last year!
A solution must be found, as the more false news there is, the harder it becomes to cross-check everything. The public will struggle to determine what’s real and what is a ‘hallucination’ — making it harder for real news publishers to present the truth to the masses. Everyone starts believing everything is fake, so who knows who’s telling the truth?!
Overall, there could be a solution to both AI news publishing problems. Perhaps new rulings should make it so AI can only read information from its own content/company. This would immediately stop copyright infringements and should prevent hallucinations as AI pulls information from facts and genuine sources.
Layoffs as a result of AI
There’s always been a worry that artificial intelligence will replace human roles. The fear is that those journalists whose jobs can be fulfilled by AI will lose those jobs in a bid to create an AI-generated newsroom of the future. Needless to say, this is a terrible idea. News publishers see AI as a way to supercharge productivity while saving money. They don’t understand the technology — or its possible drawbacks. As mentioned above, we’re getting many fake news stories from AI, and many AI models often breach copyright laws. Firing humans and replacing them with artificial intelligence is a recipe for disaster in the long run.
Instead, a focus should be on training journalists to use AI so they can improve their processes. When used correctly, artificial intelligence can be a powerful tool for news publishers. It can help develop better and more clickable headlines to drive traffic or offer insights into what types of stories are generating interest at the moment. Rather than opting for humans or AI, publishers should mold the two for long-term success.
How can News Publications work with AI?
We’re already seeing news publications strike huge partnerships with AI. Most notably, OpenAI has gone on the offensive and secured deals with Conde Nast (the owners of Vogue, GQ, and other global magazines) to use their data and content to train current and future OpenAI models.
This will become the trend in the coming years, but part of the deal also means ChatGPT can present content from these brands on its platform. Therein lies the problem for news publishers; working with AI will mean news sites lose traffic. People will use AI to find information rather than go to news sites, which translates to less ad revenue, subscriptions, and money loss.
However, this doesn’t mean news publishers should avoid working with AI. Forming partnerships with AI firms can be beneficial — but it’s important to know what you’re getting from the deal. News publishers should work with AI when they’re compensated enough to coverfor the loss of revenue seen by AI sites stealing traffic and revenue from the news websites.
Conclusion
In conclusion, AI is here to stay, and there’s not much news publishers can do to stop this. The biggest problems are copyright infringements, fake AI news stories, and layoffs. AI feels threatening for the news industry, but that doesn’t have to be the case. News publishers can (and should) embrace this technology and learn how to use it to benefit their business. Teach journalists to harness the power of AI for better headlines and stories — but also, don’t be afraid to partner with AI firms and get fairly compensated for your work with them. AI and news will go hand-in-hand as this technology progresses, and it’s all about learning to work together rather than against one another.
Looking for a platform to manage your ads inventory? Reach out to a member of our team today.
Originally posted on ADvendio.com on 29th August 2024.
