Cyabra: Israeli TruthTech Startup Examines Consumer Conversations for Authenticity

For our Startup Spotlights, Times of Entrepreneurship scouted the most successful startups that spun out of top university competitions and programs. The impact of such awards can be many years in the making, especially in the DeepTech space.

Interviewee: Dan Brahmy is the co-founder and CEO of Cyabra, a U.S. GSA certified SaaS platform that uses AI to measure impact and authenticity within online conversations.

What is your elevator pitch? Has it changed much since you started?

Cyabra measures impact and authenticity within consumer conversations by building the first filtering mechanism for enterprises and the public sector. While our elevator pitch might evolve through the years, our vision has stayed the same: we strive to be a pioneer in TruthTech, by building the best filtering mechanism available to consumer brands, agencies and public sector entities.

What should we know about you?

The co-founding team has gathered the most unique and relevant background in information warfare within the Israeli Defense Forces (IDF). Cyabra’s platform uses a semi-supervised, machine learning algorithm. We began with a static list of parameters, some unique to platforms and some broad, which defined social media fictitious behavior. Over time, this process became more automated, as the algorithm learned from itself, defining new red and green flags in account proliferation and behavior. With this approach, we can measure authenticity of authors and content as well as the impact of their efforts, including how information and sentiment, both authentic and inauthentic, can move down the ladder to skew public opinion.

What are you looking for?

We are hiring the best R&D talent in Israel to enhance our Machine Learning and natural language processing (NLP) capabilities. We are also always looking to expand our client roster to introduce companies and agencies across different industries to the value of disinformation tracking and analysis. Recently, we’ve been exploring opportunities to work with the public sector and U.S. federal agencies, as we became the first social media disinformation tracking tool available to these groups through the GSA Schedule earlier this year.

Why should someone invest in you?

We are solving a problem that affects everyone on this planet, identifying and analyzing false and misleading narratives in online content. As the technologies nefarious agents and foreign adversaries use to disseminate disinformation evolve, similar platforms using AI and ML to track and act against these efforts will become just as prevalent. We believe the impact Cyabra’s efforts will create over time is hugely valuable both for individuals as well as enterprises, worldwide.

How much total funding have you raised so far and from which sources?

Cyabra has raised $1M to date from Tel Aviv University Ventures and tech executive Sonny Vu. We’re continuing to actively fundraise and hope to announce a new round later this year.

How many employees work for your startup?

The Cyabra team currently consists of 18 employees.

Is there a clear evidence of success you would like to share?

Cyabra has put together several case studies showcasing the value and functionality of our platform across various industries. One of our most impactful looked at a tweet by Zhao Lijian, a Chinese foreign ministry official and spokesman, which included a fake photo of an Australian soldier committing violent acts. This analysis revealed an orchestrated disinformation campaign. Of the nearly 1,500 profiles that engaged with Lijian’s tweet, Cyabra found 57.5% of the accounts to be fake, working together to spread the harmful narrative published by Lijian. These profiles were flagged by Cyabra’s technology for various reasons including low quality of profile. Many accounts were also found to be created around the same time, all with just one post – a retweet of Lijian’s original tweet.

Worst and best advice received so far? 

The best advice I’ve ever received is to under-promise and over-deliver! As for worst advice, I was once given the suggestion to build multiple solutions for multiple sectors simultaneously…this isn’t the correct approach, nor is it relevant for startups. A startup should have one single north star and, often, a certain limited bandwidth to acknowledge.