As artificial intelligence becomes more and more a part of our lives, researchers are facing the challenge of climbing out of the uncanny valley.
The uncanny valley, for those unfamiliar, is the phenomenon where a digitally-generated image gets closer and closer to human-like, humans become more and more uneasy with it. Something that doesn’t look human at all, like Wall-E in the chart below, is completely comfortable for humans. But as it gets closer to human, we instantly get uncomfortable. That, in a nutshell, is the challenge with AI-generated images today.
A Los Angeles-based startup, Channel 1, is trying to overcome this challenge for news. Late last year they posted a video to Twitter (X) that showed how human-like their AI-generated hosts had become.
Channel 1 is not alone. This article from the BBC profiles efforts around the world from Kuwait, Greece, South Korea, Taiwan and India.
The article also grapples with the bigger question. With trust in news at a low ebb – just 42% of people in the UK trust newsreaders, down 16 percentage points in one year – how will people trust AI-generated news readers.
Like it or not, this seems to be the future of news. The news business model is already straining under the weight of its cost base, with layoff-after-layoff in most traditional news outlets. Moving to AI-generated hosts will be a saving to tempting to ignore, at least for a percentage of media executives. What is harder to quantify, but ultimately more important, is whether these media organisations can maintain public trust. Otherwise the move to embrace AI will be the next step in the traditional media death spiral (lose viewers -> cut costs -> lose trust -> lose viewers -> cut costs etc.).
This is an excerpt from our Thought Starters email. Once a week we send you 5 interesting articles that have caught our attention, to get you thinking. No spam, we guarantee.