Data Natives Unlimited – Europe’s biggest data science and AI event – was forced out of its regular, “so Berlin” home this year thanks to the Covid-19 pandemic. What followed was an endeavor that exceeded our expectations.
With 5,000 attendees, over 150 speakers, incredible partners, fantastic volunteers, and a team that put their heart and soul into the event, we set out to create something unique; an online conference that felt every part as good as the offline events of last year.
“2020 has been the year where an unprecedented event changed our lives, and we had to change the way we work, live, and communicate,” CEO and founder at Dataconomy and Data Natives, Elena Poughia, said. “Digital transformation happened faster and accelerated more than ever expected, and similarly Data Natives Unlimited was the experience that we put together as a reaction to the situation.”
And the journey didn’t just start with the conference.
“It was more than an event,” Poughia said. “It was an experience that started from September, with a hackathon, which we found was the best way to find quick, innovative solutions to societal problems and challenges, and then it continued with discussion roundtables tackling important topics before coming to an end with the DN Unlimited conference.”
And while we’re a little biased, we think we achieved exactly what we set out to do. Over three days, we brought you keynotes, discussions, networking, community features, and exclusive access that came as close to an in-person event as is possible across a screen.
Opening the event on day one, Poughia set the scene with a talk that sparked some interesting discussions. Looking at a year we lived online, for the most part, Poughia’s keynote turned to the privacy, security, and transparency of our data, a commodity that is now more valuable than ever.
“The world produced an immense amount of data over the past months. It is our responsibility to handle this data with care – staying both private and transparent, sharing our data while protecting it, and always keeping in mind that impact is the new money,” Poughia said.
Chris Wiggins of the New York Times then explained the ins and outs of data-driven journalism and how the world’s oldest newspaper became the forerunner in the media industry by developing a data strategy for its core activities.
Speaking of news, two themes emerged across the first two days of DN Unlimited. Fake news and deep fakes; both of great concern to many.
Juan Carlos Medina Serrano presented his research on Tik Tok as the new gatekeeper of political information and social media’s power to increase societal polarisation. Weifeng Zhong fascinated us by NLP’s applications, predicting the next major political moves by analyzing propaganda content. And we heard from the likes of Thorsten Dittmar, Kathrin Steinbichler, and Alexandra Garatzogianni on deep fakes and fake news too.
“Based on the lessons learned in the last five years of tackling fake news, we can design the proper policy response to deepfakes, but we need to spot the risks early on,” said Areeq Chowdhury of WebRoots Democracy.
Of course, the battle for the presidency in the United States made its way into the conversation, especially given it was happening at the same time as the conference.
“Take the US election; 98% percent of disinformation didn’t require any AI at all to create a very compelling conspiracy. We need to learn to think critically about the media messages we are receiving,” added Kathryn Harrison of FixFake.
There is some hope for the future, however, as we start to see more data sovereignty solutions come to fruition worldwide.
“We are entering the third phase of internet development, where citizens create control over their data. This year is all about personal privacy and internet reliability,” said John Graham-Cumming
In addition to the discussions around how our data is used, we dove deep into another topic that keeps AI and machine learning advocates awake at night; ethics and data bias.
The opening keynote from Mia Shah-Dand, CEO at Lighthouse3 and Founder at Women in AI Ethics, talked about the crisis of ethics and diversity in AI.
“We can draw a direct line between a lack of diversity and AI bias. The questions we should ask ourselves before implementing algorithms are who participated, who was harmed, and who benefitted from our solutions,” Shah-Dand said.
Listening to Jessica Graves of Sefleuria, we realized that there is no technical reason algorithms can’t eventually learn to generate creative output if we give them access to the same inputs and feedback as humans.
The conversation around data also extended to governmental and regulatory policy. Anu Bradford, Alexander Juengling, Sebastien Toupy joined us on the main stage to talk about the “Brussels effect” – the EU’s unique power to influence global corporations and set the rules of the game while acting alone. We found out that neither a ‘hard’ or ‘soft’ Brexit will liberate the UK from the EU’s regulatory reach.
One of our main partners, IBM, had a stage that was booming with insightful talks. Noel Yuhanna and Kip Yego discussed how trust in AI should start with the concrete data foundation, and Jennifer Sukis and Dr. Robin Langerak walked us through the AI lifecycle.
“I haven’t been prouder than before when it comes to our content,” Poughia said. “We really managed to get very high level quality speakers, focused on a lot of interesting topics such as the pandemic, data monopolies, deep fakes, fake news, and other areas that impact our lives.”
In addition to the conference content, we announced the winners of the DN Unlimited Hackathon we ran in September. Three winners created solutions to help bring adaptive learning to places where internet connectivity is weak, to accelerate precision medicine, and to help people measure their environmental impact.
Our EUvsVirus colleagues Michael Ionita, Urska Jez & Jesus del Valle concluded that realizing our wildest dreams is possible not only through hackathons but entirely online. The trend is here to stay.
Speaking of online, our attendees took full advantage of speaker Ask Me Anything (AMA) sessions, where participants could meet and greet our distinguished experts. Our Slack channels were buzzing with activity, and the various networking tools on offer helped to connect the masses.
“I’m glad we had a way to bring everyone together and to be connected on different mediums and formats,” Poughia said. “And it’s ‘DN Unlimited’ because there was no limit to the communication, and the connections that were made. We really were crossing all borders.”
We would love to bring Data Natives back in an offline capacity for 2021, and pandemic permitting, we’ll make that a reality. And while COVID-19 forced our hand, we couldn’t have wished for a better online event this year.
Thank you to everyone that participated, in whatever capacity – you made it special. We’ll see you all, in either two or three dimensions, next year.