AI assistants for caregivers looking after family
Which AI assistants can keep appointments and reminders, phone on your behalf, nudge you unprompted and remember medication and preferences?
23 assistants ranked
Twenty-three assistants took at least three of the four tests that matter when you look after an older relative or a family member. Instinct leads at 9.3 over three tests, then szn at 8.7 over three, Neo at 8.5 over all four, and Wajo / Fo and Poke at 8.0. Neo and Wajo / Fo are the highest scorers that took all four, and 18 of the 23 took the call test.
Top picks
- 19.3InstinctTop pick
Tested on 3 of 4Best: Routines10, Memory10
- 28.7
Tested on 3 of 4Best: Phone calls10, Memory9Time to finish 7m 30s
- 38.5
Tested on 4 of 4Best: Phone calls9, Proactive9
- 48.0
Tested on 4 of 4Best: Phone calls9, Proactive9
- 58.0
Tested on 3 of 4Best: Routines8, Proactive8
Top pick: highest average here. Best free option: highest among assistants you can use for free. Fastest that's still good: quickest median time to finish among those averaging 7 or more.
Full ranking
| # | Assistant | Routines | Phone calls | Proactive | Memory | Average |
|---|---|---|---|---|---|---|
| 1 | Instinct | 10 | — | 8 | 10 | 9.3 |
| 2 | szn | — | 10 | 7 | 9 | 8.7 |
| 3 | Neo | 8 | 9 | 9 | 8 | 8.5 |
| 4 | Wajo / Fo | 7 | 9 | 9 | 7 | 8.0 |
| 5 | Poke | 8 | — | 8 | 8 | 8.0 |
| 6 | Norton Family Assistant | 8 | 8 | 6 | 7 | 7.3 |
| 7 | Waldo | 9 | 6 | 5 | 8 | 7.0 |
| 8 | Luke | 7 | 8 | — | 6 | 7.0 |
| 9 | Vellum | 8 | — | 6 | 7 | 7.0 |
| 10 | Ask Ranger | 6 | 7 | 7 | 6 | 6.5 |
| 11 | Bantr | 6 | 5 | 7 | 7 | 6.3 |
| 12 | Familiar | 5 | 7 | 7 | 6 | 6.3 |
| 13 | Tressa | 8 | 4 | 7 | 5 | 6.0 |
| 14 | Friday | 6 | 5 | — | 6 | 5.7 |
| 15 | Buoy | 7 | — | 4 | 5 | 5.3 |
| 16 | Ka | 5 | 6 | — | 5 | 5.3 |
| 17 | tinyNature | 5 | — | 8 | 3 | 5.3 |
| 18 | Today | 5 | 4 | 7 | 5 | 5.3 |
| 19 | OpenClaw | 7 | 3 | 5 | 5 | 5.0 |
| 20 | AVA | 3 | 7 | — | 4 | 4.7 |
| 21 | Gemini Spark | 4 | 6 | 4 | 4 | 4.5 |
| 22 | Manus | 4 | 6 | 4 | 4 | 4.5 |
| 23 | Lucas | 5 | 4 | 4 | 5 | 4.5 |
How this report works
Included are assistants with at least three of Routines, Phone calls, Proactive help and Memory taken (catalog checked 2026-10-26), ranked by the average of the tests they took. No assistant in our catalog says it is built for caregivers, and we do not test medical advice, dosing or whether a reminder is clinically right. Check anything health-related with a professional.
How to choose
- For appointments and reminders, Routines: Instinct scores 10, Waldo 9, Neo, Poke and Norton Family Assistant 8. szn did not take it.
- For calls on your behalf, Phone calls: szn scores 10, Neo and Wajo / Fo 9, Norton Family Assistant and Luke 8. Instinct and Poke did not take the call test.
- For remembering medication times and preferences, Memory: Instinct scores 10, szn 9, Neo, Poke and Waldo 8. Proactive help is led by Neo and Wajo / Fo at 9.
Which AI assistant can help me care for an elderly parent?
Instinct leads at 9.3 over three tests, szn follows at 8.7 and Neo at 8.5 over all four. None says it is built for caregivers, and we do not test medical advice.
Which one can call a doctor's office for me?
On Phone calls, szn scores 10, Neo and Wajo / Fo 9, Norton Family Assistant and Luke 8. Eighteen of the 23 ranked assistants took that test.
Can I rely on an assistant to remember medication?
The Memory test checks it keeps what you tell it across days; Instinct scores 10 and szn 9. It is not a medical device, so keep your own pill schedule too.
What users say
The agent detected that Alaska had moved my flight 90 minutes earlier, a fact I'd completely missed, which qualified me for a full refund.
Open sourceSupposed to ping me when an email comes in and it doesn't for awhile.
Open sourceI’ve paired Fo with my creative director agent and they workshop X posts in my text group chat every morning lmao
Open sourceFrequently asked questions
Which assistant is best for this?
Instinct leads with an average of 9.3, ahead of szn and Neo.
Is there a good free option?
szn is the highest-ranked assistant with a free plan, averaging 8.7.
How did you test these assistants?
Every assistant got the same published tests, word for word, scored from three checkpoints and the number of corrections it needed. Each test is on the Dimensions page.