Getty Images has released an artificial intelligence (AI) tool for publishers that recommends the best choice of images to accompany a news story.
‘Panels by Getty Images’ uses customisable filters and a self-improving algorithm that learns how an editor selects an image and optimises its performance over time.
"The AI starts working for you as soon as you copy/paste your article into the tool,” says Getty Images senior vice president of data and insights Andrew Hamilton.
Artificial intelligence allows natural language processing, an established technique used to understand written text and interpreting or deriving insights from it.
The tool works like a picture editor — it reads the text and tries to understand what the story is about. It then offers the first round of picture suggestions based not only on individual keywords but the meaning of sentences and paragraphs.
It also allows the journalist to modify the text and add new keywords to refine the search.
With 110 million images to choose from, it may take several attempts to find the ideal picture, depending on the complexity of the article. It’s also trickier to illustrate a story where the visual narrative is not obvious, for example 'teenage depression' or 'hospital waiting times', says Hamilton.
“AI is a tool but the power is still in the editor's fingertips,” he adds.
To develop Panels, Getty Images partnered with Vizual.AI, a company specialised in the use of artificial intelligence and machine learning techniques to analyse, recommend, and optimise images for publisher content in real time.
Part of the algorithm also comes from the analysis of published articles and images associated with them, which then helped form editorial image choices.
“Publishers are under a lot of pressure,” says Hamilton. “They need to publish more and more content, publish it faster, and get more engaging imagery because there is the fight for customer's attention.”
But can AI replace picture editors in the newsrooms?
"When you think about a creative process of selecting an image to match the story, it's a much harder problem to try to solve with computers,” concludes Hamilton. “So even with this tool, it's the human picture editor that has the last word. It's not something technology can do a good enough job on.” Yet…
Free daily newsletter
- What does content verification mean in the age of deepfakes?
- Five surprising journalism disciplines all students should know about
- Getty Images tackles the lack of female and non-binary voices in photojournalism
- Robot journalists revive hyperlocal communities left behind by declining regional media
- Trust in news, membership models and Gen Z: here is your weekly journalism update