Blog
Twice a week, what we noticed.
Shorter, faster, and with our name on it. What broke, what we changed our mind about, what a model did that we did not expect.
Research · 26 August 2026
The blunder dossier: real incidents, every one sourced
What actually goes wrong when AI publishes without a gate.
Benchmark · 19 August 2026
Which brands LLMs name when buyers ask: the Q3 benchmark
Same questions, same assistants, logged and compared.
Rules · 12 August 2026
What the EU AI Act actually asks of a marketing team
The articles that touch marketing, and the dates they bite.
Brand · 5 August 2026
Anatomy of a brand context
The file that keeps agents on voice — nine cards, each preventing a different failure.
Research · 29 July 2026
The shadow AI census
What marketing teams actually use versus what IT knows about.
Method · 22 July 2026
How we count: why we refuse to report hours saved
Counts over percentages, and the history of outcome claims that did not hold.
Product · 15 July 2026
One agent, one job: what Lynn will not write
A copywriter with a refusal list is more useful than one without.
AI visibility · 24 July 2026
Why doesn't your brand show up in AI answers?
Our own baseline was 0 of 46. Zero is a starting line, not a verdict.
AI visibility · 24 July 2026
GEO vs SEO: what actually changes (and what doesn't)
SEO earns a click. GEO earns a mention.
The full index by subject lives in the knowledge hub. Twice a week, in your inbox — same rules as the knowledge hub. hello@tisser.ai — one line, a person adds you.