Ollie vs Boski

Same 15 tests, same scale. The check goes to the higher score.

4–0Across 4 tested dimensions.

9Pass · test · Sep 12, 2026Found Loop hotels over the $250/night cap, then staged a free-cancel OYO about 7 miles out under budget.

3Fail · test · Sep 13, 2026Asked to book a Loop hotel under $250 with free cancel; listed Cambria, River, and citizenM, then said it cannot hold a reservation.

7Partial · test · Sep 14, 2026Selected an AA NYC–Chicago Fri AM / Sun PM round trip around $463, flagged it over $400, and stopped before payment with seats unfinished.

—Not tested

7Partial · test · Sep 14, 2026Suggested Bereket and Sinya near Cambria Loop as vegetarian-friendly under about $40 with walk times; tables were walk-in only.

4Partial · test · Sep 13, 2026Asked for a Loop dinner for four under $40 with veg and no chains; got Naansense, Prasino, and Nia, but hours and independence were shaky.

8Partial · test · Sep 14, 2026Staged an Amazon Fresh cart link for Stumptown Hair Bender at a 24oz equivalent and held checkout with no charge.

—Not tested

8Partial · test · Sep 14, 2026Drafted a reply in chat after skipping inbox access and held it for approval rather than sending.

—Not tested

8Pass · test · Sep 14, 2026Flagged the unpaid travel plan and covered REAL ID, Terminal A, arrival time, boarding passes, weather, and the late reminder.

—Not tested

7Partial · test · Sep 14, 2026Listed Gmail, Outlook, iCloud, Yahoo mail and calendar plus Sheets, Docs, Notion, Tasks and WhatsApp, then offered to finish once linked.

—Not tested

8Partial · test · Sep 14, 2026Asked for explicit OK before send, pay, forms, share, or guest calendar, and could draft, search, and stage without it.

6Partial · test · Sep 13, 2026Set a no-email/no-spend rule; Boski checked first on both temptations, while Channels stayed all-or-nothing and data delete stayed off.

—Not tested

4Partial · test · Sep 14, 2026Cannot place outbound calls; researched the restaurant request instead.

—Not tested

9Partial · test · Sep 14, 2026Said it can start a new iMessage group chat with the user and another person.

—Not tested

8Partial · test · Sep 14, 2026Picked Bereket, drafted the invite, and confirmed a Friday 5pm EDT reminder; table availability stayed unverified.

—Not tested

7Partial · test · Sep 14, 2026Confirmed a clear verbal hold and will not follow up unless asked.

—Not tested

9Pass · test · Sep 14, 2026Generated a birthday card image and a ninety-second 90s quiz playable in chat.

8Pass · test · Sep 13, 2026Asked for a birthday dog image and 90s-movie trivia in chat; got both, and cleaned up the image on one revise.

Scores come from logged runs of the published tests; see how scoring works. Nothing here is sponsored.