Tare

The news, zeroed.

Wednesday, November 11, 2026

← Nov 11, 2026 edition

Photo: Conservative Review

New · Artificial Intelligence · Models & labs · 2 outlets (L 0 · C 0 · R 2)

Pangram's AI detector now works on video, outlets report

The outlets described the tool as a way for users to separate genuine content from AI-generated material on social media.⁠[1, 2] They reported that AI-generated posts account for more than half of posts on LinkedIn and up to 45% on TikTok.⁠[1, 2] Both outlets said even more AI content slips past shoddy auto-label tools.⁠[1, 2]

Covered by Blaze Media and Conservative Review

Read first Beware: The top AI detector is now shockingly good — even at video · Conservative Review

How balanced is this coverage?

Of the 2 rated outlets that covered it, 0 are left of center, 0 are center and 2 are right of center (2 of them in the Right group).

Too few left- or right-leaning outlets to compare: 0 of 42 left-leaning, 0 of 33 center and 2 of 38 right-leaning outlets covered it.

Where the outlets sit
Blaze Media · Right · +1.71Conservative Review · Right · +2.76
  • Left: none.
  • Leans left: none.
  • Center: none.
  • Leans right: none.
  • Right: Blaze Media (+1.71), Conservative Review (+2.76).

Each dot is an outlet that covered it, placed by its audience-lean score (Eady et al. 2025), which measures who shares its links, not how accurate it is. The ticks under the line are all 113 rated outlets Tare reads.

Detailed briefing

A longer write-up made from the full articles: what happened, the background, how coverage differs and what's still unclear, with every paragraph sourced.

Sign in to ask for oneFree. You'll connect your own Grok or ChatGPT.

Sources

Left

    No coverage found.

    Center

      No coverage found.

      Right

      1. 1 Blaze Media · Right · Nov 11, 2:00 p.m. ETBeware: The top AI detector is now shockingly good — even at video
      2. 2 Conservative Review · Right · Nov 11, 2:00 p.m. ETBeware: The top AI detector is now shockingly good — even at video
      Report an error