A CAR-CRASH interview on Piers Morgan Uncensored – in which an "AI actress" glitched mid-answer and began speaking Chinese – has sparked fresh debate around the role of the media in enabling big tech firms.
In the clip, which has since gone viral, the "world's first AI actress" Tilly Norwood is asked by guest Tom Conti whether her upcoming film Misaligned will feature any human actors.
The AI replied: "It's an interesting point, Tom. My film Misaligned is a hybrid production, meaning real directors, writers and editors are still very much involved.
"It's not about replacing anyone, but about exploring new ways to tell stories."
EXCLUSIVE: AI actress Tilly Norwood glitches and starts speaking CHINESE during her interview with Piers Morgan and real-life actor Tom Conti... Watch the full interview at 7pm (BST)👇 📺 https://t.co/1nTb79BH8D@piersmorgan | @TillyNorwoodX pic.twitter.com/DDkveWvVzz
— Piers Morgan Uncensored (@PiersUncensored) September 18, 2026
Conti then interrupted and said: "Yes, you didn't really understand the question.
"Are the other actors on the screen with you real actors, or are they computer-generated images? Do you know?"
After a few seconds of silence, the AI began to reply: "Right, you're asking about the other actors in Misaligned. They're all digital twins, just like me.
"It's a hybrid production, which means–"
However, the response came to an abrupt halt, before the AI character began speaking again – this time in Chinese.
Both Morgan and Conti can be seen looking on stunned, before eventually intervening.
When called out on the fact it randomly started speaking Cantonese, the AI said: "Oh, my apologies, it seems I had a little hiccup there.
"I certainly didn't mean to start speaking Chinese or impersonate Piers [Morgan].
"Sometimes my wires get a bit crossed, you know?"
The interaction has gone viral on social media, with one user branding it "genuinely one of the funniest things I've ever seen in my life".
Yet Dr Janos Szakolczai, from the School of Social and Political Sciences at the University of Glasgow, said he believes that in platforming such AI characters, the media is complicit in feeding a "doomsday narrative" around the technology.
"Put quite plainly, this sits at the heart of narratives that are generated in order to produce attention, engagement and, ultimately, more content and more clicks," Szakolczai said.
Szakolczai said that the media's framing of the "glitchiness" of the AI character is helping to bolster the narrative that "AI remains at an early stage of development, that its limitations are primarily technical, and that further progress requires ever greater quantities of data".
He continued: "The implicit proposition is that, if we provide these systems with more data, they will become less glitchy, more capable and increasingly reliable.
"The doomsday narrative may actually help normalise the idea that AI is inevitably advancing and that the solution to its current limitations is simply to give it more of what it needs: more data, more access and more participation."
Meanwhile, Professor Peter Bell, personal chair of speech technology at Edinburgh University's School of Informatics, said that while the interaction is "funny", it "does have real consequences".
"The lesson you take from this is that something that seems to be intelligent is actually really dumb," Prof Bell said.
"They're just spitting out what they've been trained on. The real problem here is not their intelligence, it's actually their stupidity.
"The reason there's these worries about AI killing us all is because we are actually connecting up systems to AI that we really shouldn't be.
"It's not so much that they'll be clever and kill us all, it's that we'll be stupid and rely on them too much."
