Incident 150: Swedish Contraceptive App, Natural Cycles, Allegedly Failed to Correctly Map Menstrual Cycle

Description: Some women using the contraceptive app, Natural Cycles, reported unwanted pregnancies, revealing its algorithm's difficulties in mapping menstrual cycles.

Tools

New Report New Response DiscoverView History

Entities

View all entities

Alleged: Natural Cycles developed and deployed an AI system, which harmed Natural Cycles users and Women.

Incident Stats

Incident ID

150

Report Count

Incident Date

2018-07-21

Editors

Sean McGregor, Khoa Lam

Applied Taxonomies

GMF, MIT

MIT Taxonomy Classifications

Machine-Classified

Taxonomy Details

Risk Subdomain

7.3. Lack of capability or robustness

Risk Domain

AI system safety, failures, and limitations

Entity

Timing

Post-deployment

Intent

Unintentional

Incident Reports

Reports Timeline

The first app to get approved as birth-control is under fire after dozens of women got pregnant — but the founder isn't surprised

businessinsider.com

‘I felt colossally naive’: the backlash against the birth control app

theguardian.com

Everything you need to know about Natural Cycles, the FDA-approved app-based contraception

glamourmagazine.co.uk

businessinsider.com · 2018

The birth-control app Natural Cycles has come under fire in Sweden after 37 women reported getting pregnant while using it.

The app, designed by physicist couple Elina Berglund and Raoul Scherwitzl, was the world's first to get approval in …

theguardian.com · 2018

Natural Cycles was hailed as a stress-free, hormone-free contraceptive. Then women began reporting unwanted pregnancies.

Last summer I had an abortion. Statistically unremarkable, yes, but mine wasn’t because of a split condom or a missed p…

glamourmagazine.co.uk · 2021

It would be fair to say that there's a lot of opinion out there on digital birth control app Natural Cycles, which recently received FDA approval to launch 'wearable' contraceptives.

The platform, which itself has had FDA approval since 201…

Variants

A "variant" is an incident that shares the same causative factors, produces similar harms, and involves the same intelligent systems as a known AI incident. Rather than index variants as entirely separate incidents, we list variations of incidents under the first similar incident submitted to the database. Unlike other submission types to the incident database, variants are not required to have reporting in evidence external to the Incident Database. Learn more from the research paper.