15. December 2022
AI, smart analytics, and machine learning are shaping the future of humanity. Autonomous vehicles, digital therapeutics, medical training, law enforcement, scientific research… The list of uses is endless.
Innovation powered by data
Yet to work effectively and drive the innovation that could transform everyday life for the better, these technologies rely on generating, analyzing, and storing copious amounts of high-quality image and video data. And there’s the dilemma…
At the same time, personal privacy is becoming more tightly controlled than ever. CCPA in the US. PIPL in China. And most stringent of all, GDPR in (and outside) the EU.
GDPR raises the privacy bar
If you think the General Data Protection Regulation (GDPR) is simply a legal framework for protecting personal data, well, think again. Because those 88 pages of requirements represent the toughest privacy and security law in the world. And while the European Union (EU) may have drafted and passed the GDPR, it imposes legal obligations onto any organization that collects data relating to EU citizens, anywhere.
So, we have a world in which innovation is powered by video, which cannot avoid capturing identities whether in the form of people, vehicle license plates, street addresses etc. And at the same time, laws that protect people’s right to privacy are being strengthened and extended around the world.
The threat to innovation
This means that, with the growing capabilities of facial recognition technology, public video data collection can pose a significant challenge to any organization that uses data. So how do we pursue innovation while remaining compliant with such exacting standards? Or have we reached the point where privacy regulations threaten to brake – or even block – innovation?
Why traditional responses just don’t work
Anonymization techniques such as pixelation and black bars have long been the traditional response to this dilemma. Even so, they cannot preserve the accuracy and integrity of the original data. Since data quality represents the backbone of AI innovation and machine learning, the result is a trade-off between privacy and video analytics.
Changing the game
Deep Natural Anonymization (DNAT), a generative AI-based technology, is changing the game. DNAT is a unique privacy technology based on generative AI. It uses artificial replacements in video and images to protect individuals from recognition. So rather than blurring the subjects in question, DNAT creates synthetic face overlays for example, or replaces license plates with replicas.
Yet here’s the magic. DNAT also preserves the quality of the original data to ensure compliance to global standards. In doing so, it eliminates the compromise between privacy and innovation.
“This technology makes data collection in public compliant according to privacy regulations worldwide, such as GDPR in Europe, CSL in China and the upcoming CCPA in the US.”
The Washington Post, March 21st, 2019
How does DNAT solve the dilemma?
DNAT uses AI to automatically detect faces and other identifiable elements, such as license plates, in the original video footage or imagery. It then randomly generates artificial replacements that reflect the original attributes. These non-reversible overlays are then applied to the original, ensuring that re-identification by facial recognition technology is impossible.
The benefits of DNAT
Apart from allowing organizations to use data in compliance with all relevant standards, DNAT is safe. The re-identification of subjects using facial recognition technology is simply impossible. DNAT is also highly accurate. Age, gender, race, emotions, facing direction, intention are all retained for analysis and AI development. In fact, segmentation maps are almost identical compared to the original. And above all, this method of anonymization is fully compliant, holding EuroPriSe certification for privacy-compliant IT products.
Accelerating innovation in automotive
DNAT is already out there, helping organizations across a broad range of sectors to safely pursue innovation in accordance with privacy law. Thanks to DNAT, for example, automotive companies can collect the data required for training machine learning models, such as autonomous driving, without compromising data quality.
Operating safely in healthcare
The medical sector has a great need to anonymize the identities of both patients and staff in sensitive medical video and image data. DNAT is widely used across a wide range of healthcare applications, from user experience and machine testing to digital therapeutics and patient tracking.
Keeping an eye on the public sector
DNAT also enables the privacy-compliant use of intelligent video analytics and data storage in the private sector. Deutsche Bahn, the largest transportation provider in Germany, is a notable example. Thanks to DNAT, DB used existing camera infrastructure for intelligent video analytics in trains and stations – as well as the development of autonomous trains.
Powering innovation. Protecting identities.
According to Booz & Co, innovative organizations boost revenues by 11%. Yet bitcom.org estimates that 90% of companies have had to put innovative projects on hold because of data protection requirements. And three quarters say that the concrete requirements of the GDPR have directly resulted in the failure of innovation projects.
Data has become an integral driver of innovation. At the same time, increasing robust and expansive privacy laws place tight restrictions on how this data can be used. Conventional responses lack the capability to preserve data quality, threatening to slow down or even stop progress. The next generation of anonymization has solved this dilemma, ending the trade-off between privacy and video analytics and freeing companies to innovate safely and responsibly.
Meet brighter AI. Powering innovation. Protecting identities.
brighter AI represents the only certified value-preserving video redaction software to assure GDPR compliance. In doing so, we end the trade-off between privacy and video analytics. If you’d like to see a demo, or simply have a chat about data compliance, just get in touch.
If you’d like to learn more about DNAT, check out this whitepaper for more information.