The document discusses the challenges of ethics in data-driven marketing, highlighting how companies use customer data to target demographics, particularly expectant parents. It addresses issues with algorithmic biases in image tagging and the legal implications of using genealogy databases like gedmatch for criminal investigations. Additionally, it touches on the controversy of drug pricing and how fitness tracking data influences insurance discounts.

1. Alex Meadows
TDWI Carolinas Meetup – October 26 2018
Ethics In a Data Driven World

2. 2
● General warning – The examples shared can be very
unsettling and uncomfortable.
● Trying to keep bias out of these conversations can be difficult,
especially when talking actual examples.
● Let’s do our best 
● Also, while an answer may be obvious to you – it may be
slightly/completely different to others, please respect that

3. 3

4. 4
● Target (and most companies) assigns every
customer a Guest ID
- loyalty/debit/credit card
- Email
- Name
● Tracks every purchase ever made
● Attaches any demographic information known
● New parents are a gold mine
- Don’t care as much about brand loyalty
- Frazzled and overwhelmed
● Birth records are public
- Everyone markets to parents upon birth
● How can companies market to expectant
parents in first/second trimester?

5. 5
“[Pole] ran test after test, analyzing the data, and before
long some useful patterns emerged. Lotions, for
example. Lots of people buy lotion, but one of Pole’s
colleagues noticed that women on the baby registry
were buying larger quantities of unscented lotion
around the beginning of their second trimester.
Another analyst noted that sometime in the first 20
weeks, pregnant women loaded up on supplements
like calcium, magnesium and zinc. Many shoppers
purchase soap and cotton balls, but when someone
suddenly starts buying lots of scent-free soap and
extra-big bags of cotton balls, in addition to hand
sanitizers and washcloths, it signals they could be
getting close to their delivery date.”

6. 6

7. 7

8. 8

9. 9

10. 10
"We’re appalled and genuinely sorry that this
happened. We are taking immediate action to
prevent this type of result from appearing. There
is still clearly a lot of work to do with automatic
image labeling, and we’re looking at how we can
prevent these types of mistakes from happening
in the future.“ – Google rep to Ars Technica

11. 11
But computer algorithms aren't perfect, and when
they identify images incorrectly, the results can
be disastrous.
Some concentration camp photos received
inappropriate tags, including "sport" and "jungle
gym."
Flickr had also been tagging some images of
people as "ape" and "animal," including a photo
of a black man named William taken by
photographer Corey Deshon, according to the
Guardian.
The photo service had also labeled a white
woman wearing face paint as "ape" and "animal,"
so Flickr's algorithm does not appear to be taking
a person's skin color into consideration when
auto-tagging them.

12. 12

13. 13
● GEDMatch – free/open
genealogy genetics database
● Takes data from all major DNA
testing companies
● Allows cross-analysis
● Also can catch criminals

14. 14
● DNA from over 100 crime scenes uploaded
● Police create fake profiles, upload suspect data and triangulate
● Forced GEDMatch to change their Terms of Service to allow for criminal investigation
● Users have mixed reactions
- A few deleted their data
- Others were thankful
• [GEDMatch’s co-creator] says a woman wrote that her father was a serial killer, and she
wanted her data out there to give the families of his victim’s closure.
● Police also have CODIS – the government criminal DNA database
- Only looks at small snippets of DNA – 20 locations in human genome
- DNA Tests/GEDMatch contains full DNA sequences – 600,000 locations in human genome

15. 15
● Problems
- Is it legal? Yes, and admissible.
- Accuracy?
• Officers could use this evidence as ‘smoking gun’
• Need to strengthen case with other evidence
- How about those hidden family secrets?
• Children up for adoption
• Not the biological parent situations
• Do ‘lost’ family members want to be found?
- Who owns the data after the donor dies?

16. 16

17. 17
● Many insurance companies are starting to offer fitness tracking discounts
● Subscribers provide fitness tracking data
● Insurance company provides discounts based on activity
● Same model used in car companies
● Problem:
- If health issue found, what are the ramifications?
• Cut policy?
• Raise rates?
- If pre-existing condition?
- What if subscriber cheats?

18. 18

19. 19
● Mylan bought rights to Epipen in 2016 (originally invented in the 1970s).
● Pre-Mylan, Epipens cost $57
● Post-Mylan, Epipens cost $600 - this was a data driven decision
- Also created a generic for $300 – this was a public relations decision
● Problem:
- If market demand will support the higher cost, is that okay?
- What about folks who can’t afford even the generic?
- Could this impact other medicine costs?

20. 20

21. 21
One of the ads, called "Supermarket," also
fools with a stereotype: It shows a father at
a market with his daughter getting the
ingredients for a traditional recipe, three-milk
cake (generally made with condensed, fresh
and evaporated milk). Latino men are
typically not depicted in such a place. So
why show a man in a market?
"Because Hispanic women love it," said Jeff
Manning, the executive director of the milk
board who worked for 25 years in the
advertising industry before taking the post in
1993.
"They love the idea. It is aspirational. They
look at it and say, 'That makes me feel
good,' " said Manning.