
Our Latest Research
From meta-analyses across dozens of existing studies to original experiments we run ourselves, here's what's come out of the lab.
Apr 27, 2026 • 30 mins read
How Consistent Are LLM Recommendations? We Tested 4 Models Across 12,000 Runs
We asked the same three prompts to four LLMs — 12,000 times in total.What we found: rankings aren’t fixed, models behave very differently, and consistency depends on more than you’d expect.

Usman Akram

Apr 27, 2026 • 30 mins read
Optimizing for AI Search Engines: A Meta-Analysis of 19 Research Studies

Usman Akram
