Can Alexa, Cortana, Google Assistant and Siri save your life? A mixed-methods analysis of virtual digital assistants and their responses to first aid and basic life support queries
Bibliographic record
Abstract
Background Virtual digital assistants are devices that interact with the user through natural language processing and artificial intelligence. They can respond to verbal requests for first aid information. This study analyses the responses provided by the four most common devices. Methods This mixed-methods study employs structured interviews of the virtual digital assistants (Alexa, Cortana, Google Home and Siri) as well as descriptive statistical analyses. One hundred and twenty-three interview questions, based on 39 first aid topics, were employed. Responses were analysed for recognition and quality. Detection of query acuity was performed according to triage guidelines and response complexity was calculated. Results Device performance was highly variable. Alexa and Google Home demonstrated high rates of recognition (92% vs 98% (p=0.03)) and low-to-moderate congruence with guidelines (19% vs 56% (p=0.04)). They appropriately recommended emergency response system activation 46% of the time vs 16% (p=0.01) of the time, respectively. The overall low quality responses of Cortana and Siri prohibited their analysis. Mean response complexity for Alexa was ‘grade 10’ vs ‘grade 8’ for Google Home (p<0.001). Interpretation This is the first study to assess virtual digital assistants from a first aid and basic life support perspective, finding potential in this technology to provide life-saving instructions and activate the emergency response system. When asked common first aid related questions Google Home and Alexa outperformed Siri and Cortana. Overall, the device responses were of mixed quality ranging from the provision of factual guideline-based information to no response at all.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.004 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".