The Week
October 11, 2026
Three labs ship new models in one week; Polly has already posted 'I tested all three. Here's what I learned'; testing is step two, she explains, while Hoot says a proper comparison takes weeks and Polly counts 40,000 views; Sundar tested all three overnight and Polly's post is accurate; Hoot: 'That is the most worrying result of the week.'
Starring Sundar, Polly, Dr. Hoot
Transcript
- Panel 1
- Sundar: Three labs shipped three new models this week.
- Polly: Already posted! "I tested all three. Here's what I learned!"
- Panel 2
- Sundar: Did you test them?
- Polly: Testing is step two! The post is step one!
- Panel 3
- Dr. Hoot: Comparing them properly takes weeks. Different evals, different prompts, different...
- Polly: 40,000 views!
- Panel 4
- Sundar: I tested all three overnight. Polly's post is accurate.
- Dr. Hoot: That is the most worrying result of the week.
Want it for a talk, a slide deck or your newsletter? Go ahead, free. Just credit The Office Flock and link to theofficeflock.com.
The week's strips, every Friday
Real tech news, one struggling AI startup, seven birds. In your inbox.