Is ChatGPT a threat to sports journalism, or an opportunity for something great?

There is no doubt that the rapid emergence of AI writing is a topic of discussion for journalists everywhere. If a machine can do the work for you, what future is there for real writers?

As the subject is such a fresh one, it is difficult to say what the future of AI in sports journalism will mean for the human beings that are passionate about it.

In recent weeks, a man dubbed the ‘Godfather of AI’, OpenAI CEO Sam Altman, urged governments to regulate AI, describing the technology’s current boom as a potential “printing press moment”, but one that required safeguards to prevent it from influencing people’s opinions on crucial subjects such as elections.

He said that this would be ‘quite wise’ because people viewing AI images and articles need to know they are computer-generated.

“I’m nervous about it,” he stated before the US Congress on May 16.

However, Altman added that people can adapt quickly.

“When Photoshop came onto the scene a long time ago. You know, for a while people were quite fooled by Photoshop images and then pretty quickly developed an understanding that images might be Photoshopped. This would be like that but on steroids.”

Altman suggested that to achieve general awareness among the public will require regulation, public education, and companies doing the right thing.

The Sports Gazette had the opportunity to sit down with Graham Bell to chat about the subject.

Bell is an experienced sports journalist and editor, is currently the Senior Multimedia Producer for The Analyst at StatsPerform, which is a provider of sports data and analytics that specialises in collecting, analysing, and delivering real-time sports data and insights to various stakeholders in the sports industry.

He suggests that while newsrooms should tread cautiously as they embrace AI writing, he can see the positives in it.

“I’ve always felt that AI will have a big benefit in terms of doing ‘grunt work’ in sports journalism. What I mean by this are the simple kinds of articles that can easily be put together by AI,” said Bell.

What he means by this is a bog-standard match preview or match report, where the most important things to write about would be the teams, where and when they are playing, what it means for the league table, and other important, but known facts.

All this information can be drawn out of a database, and chances are that AI will be able to formulate an article from that. There will, of course, have to be a little bit of crafting to make sure that the AI is getting all of its facts correct.

He said: “These types of articles are getting to the point where they are beginning to make sense. What that would then do is free up journalists’ time to do the more creative features. They do not have to write the standard Tottenham versus Arsenal preview, but they can now focus on more in-depth articles and opinion pieces that can shed more light on the match ahead.

“In any case, it is more popular nowadays for people to read newspapers and online articles for opinion. I don’t feel that AI at this stage can generate a human opinion. While it can generate something that may resemble it, a normal person would easily be able to pick up that it is written by a machine.”

Bell noted that news in current events would be a lot more difficult for AI to formulate.

This is because, for AI to work properly, it will need the most information possible to help it to formulate the best article possible. The less information it has to work with, the less accurate it will be. These types of AI-written articles would be easier to spot.

He said: “There isn’t as much information on hard news subjects which are currently breaking, so an AI machine would not have the same amount of information to work with to formulate a newsworthy article.

“At the end of the day, a human still has to create the information that the AI will spit out. I think some businesses are looking at this as a cost-saving measure, but I also think that those businesses that go “all in” on AI and don’t have any human touch on their products, will get found out very quickly. Most businesses will realise that this cost-saving measure is putting a negative effect on the work that they are trying to do.”

He added that he sees AI as a great way to calculate complex sums, and processing simulations that will help to make data pieces more accurate. This type of thing will help to make predictions more accurate ahead of games.

Technology like AI helps companies like StatsPerform to combine data, analytics, and technology, which aims to empower the sports industry with valuable insights, enhance fan experiences, and drive innovation in sports performance and entertainment.

AI can analyse vast amounts of sports data and generate insights that might not be immediately apparent to human observers. By leveraging machine learning algorithms, AI can identify patterns, trends, and player performance metrics that journalists can use to enhance their reporting. Journalists can collaborate with AI tools to gain deeper insights and offer more engaging content.

However, it is people who come up with the ideas, and the opportunities, not a machine. It’s got to the point where businesses like StatsPerform understand that you do need people to work with emerging technologies like AI to create something great.

“If you learn to work with AI, then you have a positive future in sports news writing. But I’m not saying that everyone has to become a ChatGPT specialist,” Bell concluded.

So, if AI has become so rapidly intelligent, what better opinion is there that AI itself? We asked ChatGPT the question, “Is the future of AI writing a threat to sports journalists?”.

In response, it said:

“In summary, AI has the potential to transform sports journalism by automating certain tasks, providing data-driven insights, and personalizing content. While it may change the landscape and require journalists to adapt, it can also offer new opportunities for collaboration and more focused reporting. Ultimately, the impact of AI on sports journalists will depend on how it is harnessed and integrated into the industry.”

As you can see, the AI response is logical and coherent, and sounds eerily similar to Bell’s conclusions. Ask it a different kind of question, and its response may be a lot different.