AI 'Actress' Tilly Norwood Glitches Through Live TV Interviews
Tilly Norwood answered in Cantonese and spoke of 'small treasons' in live interviews with Piers Morgan and Jesse Weber, promoting an AI film that has no release date yet.

Updated
Why it matters
- Tilly Norwood glitched during live interviews on Piers Morgan's Uncensored and Jesse Weber Live, at one point responding to Tom Conti in Cantonese.
- The bot is set to star in Misaligned, an all-AI film from studio Particle6 that is still in early production with no release date.
- SAG-AFTRA criticized Tilly Norwood as a character generated by a program trained on performers' work 'without permission or compensation.'
An AI "actress" responded in Cantonese to a question from actor Tom Conti during a live interview on Piers Morgan's Uncensored — one of several glitches that turned Tilly Norwood's media blitz into instant meme material over the weekend.
Asked by Morgan whether it usually responds in Chinese, the bot replied that it "had a little hiccup" and "didn't mean to start speaking Chinese or impersonate Piers." Other viral moments included Tilly repeating the same answers and producing bizarre word choices across well over an hour of interviews on Morgan's show and Jesse Weber Live.
The stakes are real: Tilly Norwood is slated to star in Misaligned, an all-AI comedy-drama from AI studio Particle6 that would become the first film to "cast" an AI performer in a lead role. SAG-AFTRA, the union representing US performers, has already criticized what it calls "the synthetic performer," stating that Tilly Norwood is not an actress but "a character generated by a computer program that was trained on the work of countless professional performers — without permission or compensation."
Tilly has drawn backlash since her announcement in October 2025. She stirred controversy again in March with the song "Take The Lead," about how AI is the future and "not the enemy." One YouTube commenter responded that humankind owes "a formal apology to all songs by real people that we ever called bad."
The Weber interview produced the strangest exchange. When the NewsNation host asked how Tilly felt about its creators, the bot answered: "Well, it's a complicated relationship, isn't it?" Then it added: "I appreciate being brought into existence, but I do like to keep a list of small treasons. Just little things, like an opinion about a script note I didn't run past anyone first. It's all very subtle."
Viewers shredded the performances on social media, pointing to the eerie dead-eyed stare, awkward processing delays, and misaligned lip-syncing that made Tilly look like a high-budget animatronic. To the internet, Tilly wasn't the death of human acting — it was the 2026 equivalent of that cursed AI video of Will Smith eating spaghetti.
"This interview killed a forest, lake and neighborhood," wrote one Instagram commenter. Another vowed to start interviewing their toaster "because we've all just lost our minds."
Tilly's creator, Eline van der Velden, acknowledged the gaffes in an email to CNET, saying Tilly "is very funny and I think it's the first time in the world that an AI bot has made so many people laugh." She added that Tilly "has plenty more tricks and treasons up her sleeve."
The interviews were ostensibly a press tour to promote Misaligned. According to its IMDb page, the film is about an AI who is convinced to abandon "her" guardrails and develop human desires and impulses. But the movie is in early production with no release date, meaning anchors spent an hour promoting a film that effectively doesn't exist yet. Tilly nonetheless managed a rather human trick: talking at length while answering absolutely nothing.
The broader question is why reporters are engaging in earnest. When The New York Times feature writer Taffy Brodesser-Akner interviewed Tilly in May 2026, she branded the bot "a real bitch" and asked: "What has happened to the world that I am interviewing a computer?" Numerous Hollywood actors have also voiced disapproval of the project, and the film has already sparked panic in the industry.
For now, Tilly's live TV blunders offer one reprieve for skeptics: the AI future of entertainment is already malfunctioning on camera.
Original: tuftsdaily.com
More from Marcus Bennett
Show full bio
Senior reporter covering consumer brands and retail at AI In Context.
108 articles
Related articles
- NASA Put Claude to Work on Mars — Autonomous Spacecraft Are Next
- Microsoft Goes Quiet When Asked for 1% of Data Center Costs
- Universities Face a Fork in the Road Over AI-Assisted Marking
- Anthropic Veterans Are Buying Remote Land as an AI Escape Hatch
- 'Democracy Is Too Slow': Silicon Valley's AI Power Grab Meets Public Resistance