An interview series spotlighting global tech influencers, disruptors, visionaries, and of course, innovators.
Happy Holidays! Wishing you a healthy, safe and happy holiday season!
Cyabra, a data visualization software company, uses advanced technology to counter disinformation and root out bad actors in order to empower and protect global brands, media outlets and the public sector from fake content and profiles, manipulated images and deepfakes transmitted across the digital realm. Using machine learning infused with NLP, Cyabra also identifies behavioral patterns typical of fake profiles such as bots, trolls and sockpuppets.
Founded in 2018 by a team of world-renowned information warfare and offensive cybersecurity experts, Co-Founder and CEO Dan Brahmy leads his team to reverse engineer the technology of bad actors and study online media trends, helping global brands and large news outlets understand their audiences and counter disinformation. Their expertise provides brands and media with the resources to better understand narratives, discover trends and uncover more in-depth insight into what drives consumers, helping companies make smarter decisions.
In addition to providing social media impact and authenticity analysis, Cyabra is a pioneering leader of deepfake detection technology. The solution focuses on detecting generative adversarial networks (GAN)—a class of machine learning that bad actors use to create realistic images and videos to negatively influence the public or manipulate videos and images.
I had the opportunity to connect with Brahmy via email to discuss a variety of topics including his career, fighting disinformation and deepfakes, GAN, growth, leadership skills and hats (!). Our interview follows.
EKMH: Cyabra identifies disinformation sources and tracks their spread, protecting global brands, media outlets and election integrity. What led to your co-founding a platform that detects deepfakes and generative adversarial networks (GAN) utilizing machine learning? How did you rally your team, and later investors, to support your data visualization software company?
Dan Brahmy: Detecting deepfakes and GAN is a crucial part of our vision- to become the online filtering mechanism for brands. Our original platform accomplished this by analyzing social media campaigns to identify sources of disinformation such as bots and sockpuppets. However, when we saw the rise of the threat surrounding visual manipulation, on top of the threat of identity manipulation, we understood that we needed to stay ahead of the curve and detect deepfakes and GAN as well. We developed a semi-supervised solution with 90% accuracy to do so, on top of our existing solution.
Rallying people around Cyabra was never a big challenge since we are solving a crucial problem that is becoming more and more relevant. We are trying to bring authenticity to the world, and usually people interested in the same vision share a similar DNA.
EKMH: What’s your vision for Cyabra? What problems do you anticipate as Cyabra grows and how will address these changes?
Dan Brahmy: Our vision is to share our authenticity lens to brands worldwide and eventually become the "stamp of approval" of online interactions (similar to an SSL certificate). I believe that the problems which will arise are related to the levels "breachability" of existing and upcoming social platforms. This means a raise in social engineering (i.e falsifying your identity, distorting other people's opinions, influential campaigns orchestrations and more). Bad and fake actors will always find ways to bypass efforts to combat disinformation and thus we have to think like them and continuously develop the solutions to detect, analyze, and help remediate against these actors.
EKMH: How can you and your team of world-renowned information warfare and cybersecurity experts stay current -- and steps ahead -- to address dynamic bad actors and their morphing disinformation methods?
Dan Brahmy: Research is key here. This research is being conducted internally as well as in partnership with academic partners (stay tuned for the next ones).
EKMH: How have your past jobs prepared you for your role as CEO? What does it take to be an effective leader and “wear a lot of hats”? Favorite/ least favorite hat?
Dan Brahmy: I see that you read my LinkedIn profile 😊... My favorite hat is customer discovery. My least favorite is financials (you can't be good at everything). Honestly, nothing can prepare you for such a rollercoaster, but I'd say that the two skills that every entrepreneur (CEO or not) should always be improving are perseverance and resilience. At an early-stage startup, your job is to never give up when facing challenges, and (un)fortunately enough, you will face hundreds of them. My former startup gave me the foundation and helped me learn from my mistakes.
EKMH: Lessons learned from your most incredible failure(s)? most significant breakthrough(s)?
Dan Brahmy: Although I'm only 29, it feels like answering a one-liner regarding my failures wouldn't do them justice. I can say that a few years ago, I lost my father to pancreatic cancer which was by far one of the most difficult things to go through as both a person and as an aspiring entrepreneur. But, at the same time, this experience also brings maturity and life perspective.
EKMH: What critical advice do first-time entrepreneurs and founders need to know?
Dan Brahmy: Within the boundaries of sanity, one should not give up in the face of “No.” Do not hope to build first and assume they will come later. Rather, create small iterations with prospective, relevant stakeholders and constantly listen until your MVP becomes indispensable.
EKMH: Which books are on your nightstand and which podcasts are lined up on your smartphone?
Dan Brahmy: Zero to One: Notes on Startups, or How to Build the Future by Peter Thiel.
The StartUp Podcast by Gimlet Media.
**Follow me @crowdfunderin and @ekmhinnovators!
Search below and read more interviews in the EKMH Innovators Interview Archive or via MuckRack…Don’t miss an Interview or prediction! Be a part of the conversation.