SURVEY REPORT 2026 · MARCH

Do You Know
What's in
Your Food?

15 respondents shared their habits, frustrations, and expectations around food ingredient labels and AI-powered additive recognition tools. Data collected March 2026.

15
Valid responses
100%
Interested in AI scanner
93%
Want risk-level labels
80%
Troubled by chemical terms
01 — RESPONDENT PROFILE
Who Answered This Survey?
Gender, age, and identity breakdown of the 15 participants.
Q1 · GENDER
Q2 · AGE GROUP
Q3 · IDENTITY TAGS (MULTI-SELECT)
Health-conscious general public
12 responses · 80%
Parent / caregiver (concerned about child food safety)
5 responses · 33%
Fitness / weight management
0 responses · 0%
Allergy-prone / pregnant
0 responses · 0%
02 — LABEL READING BEHAVIOR
How Often Do People Check Ingredient Labels?
Q5 & Q6 — Frequency and purpose of checking food labels.
Q5 · LABEL CHECKING FREQUENCY
Q6 · WHY CHECK LABELS? (MULTI-SELECT)
Worried about additives (preservatives, colors, flavors)
10 · 67%
Worried about sugar / fat / calories
8 · 53%
Just curious, browsing
5 · 33%
Worried about allergens
3 · 20%
03 — PAIN POINTS
What Makes Reading Labels So Hard?
Q7 — The biggest frustrations when reading ingredient lists (multi-select).
Q7 · FRUSTRATIONS WITH INGREDIENT LABELS
Full of chemical jargon — no idea what things are
12 · 80%
Know there are additives, but don't know if they're safe or harmful
11 · 73%
Want to avoid certain ingredients but can't identify them on the label
11 · 73%
Read the label but still don't know whether to buy — no clear conclusion
8 · 53%
Font too small — straining to read
5 · 33%
Waste of time checking on the shelf — too slow
3 · 20%
80%
can't make sense of chemical terms in ingredient lists
73%
know additives are there, but can't assess their safety
53%
still don't know whether to buy even after reading the label
04 — PRODUCT NEEDS & EXPECTATIONS
What Would an Ideal AI Scanner Look Like?
Q8–Q11 & Q14 — Interest, classification preferences, feature requests, and payment model.
Q8 · INTEREST IN AI FOOD SCANNER
Q14 · PREFERRED PAYMENT MODEL
Q9 · PREFERRED CLASSIFICATION STANDARDS (MULTI-SELECT)
Risk level: green (safe) / yellow (caution) / red (high-risk)
14 · 93%
Functional type: preservative, colorant, sweetener, etc. — with explanation
13 · 87%
Plain-language name (e.g. "artificial vanilla flavor" instead of E621)
9 · 60%
Population tags: pregnant-safe, allergy-friendly, child-safe
10 · 67%
Source: natural extract / nature-identical / synthetic
8 · 53%
Q11 · ADDITIONAL FEATURES WANTED (MULTI-SELECT)
Find a "cleaner" alternative product in the same category
12 · 80%
Save scan history — personal consumption red/green list
12 · 80%
Auto-calculate total sugar per bottle in common units
11 · 73%
Identify cosmetics / skincare ingredients
10 · 67%
100%
of respondents are interested in an AI food additive scanner (40% very interested, 60% somewhat)
93%
want a traffic-light risk classification system (green / yellow / red)
100%
prefer a free model with ads over paid subscriptions
SENSE Institute · Food Additive AI App Survey Report · March 2026
15 valid responses · Lead Researcher: Zhiyan Chen · senseinstitute.org