Urgent.News

What's breaking now, across thousands of outlets.

AI

AI 'actress' suddenly starts speaking Chinese during TV interview

An interview with AI-generated “actress” Tilly Norwood took an unexpected turn when the virtual performer suddenly stopped answering in English and began speaking Chinese. The incident happened during an appearance on Piers Morgan Uncensored alongside British actor Tom Conti, as Norwood continues a media tour promoting the upcoming film Misaligned. Get updated faster and for FREE: Download the…

During a TV interview on Piers Morgan Uncensored, the AI-generated "actress" Tilly Norwood unexpectedly began speaking Chinese. The incident occurred while the virtual performer was promoting the upcoming film Misaligned alongside British actor Tom Conti. Conti questioned whether the other actors in the movie were real people or computer-generated.

Initially answering that real human writers, directors, and editors were involved, Norwood then explained that the cast were in fact "digital twins" like herself. After a brief pause, she started speaking Chinese for over ten seconds before apologizing and attributing the incident to a "little hiccup" caused by her "wires getting crossed."

Norwood, an AI-created character developed by Xicoia, the AI division of Particle6 Group, has been generating media attention since her debut in 2025.

Written by urgent.news from Gulf News's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at gulfnews.com →

More in AI

Turn chats into Skills, Skills into scripts

Here's a tip to make your agents faster, burn fewer tokens, and behave more consistently: If you find your agent repeating certain tasks, ask it to reflect on the conversation and turn it into an…

  • Convert repetitive tasks into Agent Skills to save tokens and reduce costs
  • Transform Agent Skills into scripts (Bash or Python) for faster, consistent execution
  • Balance automation benefits with effort required for maintenance

Nobody talks about RAM. Every local-LLM regret is a RAM problem.

Originally published at mrsaynothing.dev . Walk into any local-LLM thread and the arguments are about GPUs. VRAM benchmarks, 24 GB cards, CUDA versus ROCm, whether the 3060 is still the people's card.

  • RAM is the primary constraint for local-LLM performance, not GPUs.
  • A 32 GB RAM Ryzen desktop struggled with 7.0 GB reserved memory for llama3.1:8b.
  • Quantization levels are necessary to fit memory requirements within hardware limits.

More from Saturday 19 September →