Streamlining Privacy & Compliancy | Patricia Thaine & Pieter Luitjens - GRIT ED. 7 Your Next Move - Magazine - Page 20
Streamlining Privacy & Compliancy | Patricia Thaine & Pieter Luitjens
PRIVATE AI
Streamlining
PRIVACY & COMPLIANCY
This team of distinguished co-founders is building the Twilio of Privacy – making
complex privacy tools accessible to developers in only a few lines of code.
With Private AI, developers can integrate highly accurate redaction, de-identification, and
synthetic personal data generation anywhere within their workflows so they can detect or get
rid of data that will get them into trouble anywhere within their software pipelines.
When thinking about what product to bring into
market, almost anything related to combining AI with
unstructured data analysis (e.g., email analysis, SMS
analysis, customer service call analysis) would have
required these founders to build a layer of privacy
that just did not exist in the market. To use privacy
solutions, companies either needed to transfer their
users’ data to a third party, use sub-par open-source
solutions that don’t properly protect user privacy, or
build a solution in-house when developers’ time is
best spent focusing on their company’s core products
and not on privacy which very few actually have
expertise in. “We decided to focus on creating the
best products possible for developers and AI teams
who need to have the outputs of privacy-enhancing
technologies easily work for their needs.”
20
GRIT
In 2017, the team was certain that apps and web
extensions would be all over their product and that
image and video processing would be just as attractive
to the market as text. When version 1.0 of Private AI
was created to be what they thought to be a really
cool product, that, in the end, just wasn’t projected to
scale. A very academic way of going about creating a
product– instead of building a hammer around some
nails, they kept looking for nails to fit the hammer. –So,
the idea was scrapped in 2018 to focus on research
before diving back into market analysis and product
ideation in 2019.
Doubling down on text processing in the cloud and
on-premise was the move since the market wasn’t
yet ready for on-edge deployment. Private AI spent
a significant amount of development time building
good technology for image and video processing but
pivoted to the market demands and put it aside.
“What didn’t cost
much money did
cost time.”
Patricia Thaine
Next AI, 2021
Pieter Luitjens
Next AI, 2021
Legislation changing the technological capabilities
and product offerings in such a short period of time
through the GDPR really shows that what is often said
to be impossible is possible with the right pressures
and incentives.
With the European Commission working on regulation
for IoT Cybersecurity and proposing a legal framework
on AI, Private AI plans on seeing an expansion of deep
technological offerings enabling compliance showing
once again how powerfully legislation can affect
innovation.
It’s not always legislation that has to keep up
with tech: we’re definitely seeing a two-way
street.
‘‘
Patricia Thaine is the Co-Founder and CEO, a
Computer Science PhD Candidate at the University
of Toronto, and a Postgraduate Affiliate at the Vector
Institute doing research on privacy-preserving natural
and spoken language processing. Co-Founder and
CTO, Pieter Luitjens, is one of the first people to build
traffic sign recognition models from scratch for a
major German auto manufacturer, which are deployed
in millions of cars worldwide. Gerald Penn the third
Co-founder and CSO, Gerald Penn is a Professor
of Computer Science at the University of Toronto,
where he studies spoken language processing and
computational linguistics, with experience working at
Bell Labs and NASA.
ADVICE
FROM A
FOUNDER
Be kind but firm.
GRIT
21