10 January 2019

Solid as a cloud

With his new project Solid, the inventor of the World Wide Web, Tim Berners-Lee, tries to give users back ownership of their data. The internet hasn’t developed in the way he intended it. Data science progresses at a dizzying pace – are the people in the driver’s seat aware of their responsibility? We asked Gerhard Weiss, Professor of Computer Science and Artificial Intelligence and Frank Thuijsman, Professor Strategic Optimization and Data Science.

As people begin to realise what is going on, the pressure on companies to self-regulate increases. If one doesn’t want to trust that sort of thing, legal changes like the EU’s GDPR can also protect users. But according to Thuijsman, the devil’s in the detail: “It’s not like all of a person’s data is in one single file, which you can neatly delete. Developers didn’t prioritise privacy until they had to.” If it works, Solid would be a kind of data safe that users control. “Having seen how the internet has developed, Berners-Lee now tries to restore the original vision.”

Naiv-IT?

Has he been online and is now seeking redemption for the narcissistic, rapacious, bigoted, pornographic golem he has unleashed upon the world? “I don’t think anyone could have foreseen what would become of the World Wide Web back when it was ushered in,” says Weiss, “all Berners-Lee probably just wanted was to create a kind of democratic space in which people could communicate and share ideas.” Was the ruthless financial exploitation of that technology really unforeseeable? Or are does the field of IT focus on technical solutions at the expense of reflecting on the ethical implications of innovation?

The DKE professors have seen a surge in popularity – but also a shift in attitudes. “Here at UM, we’ve been researching and developing AI for 25 years now – so since before the term data science was coined. You can definitely see from the number of applicants that more and more people are interested. We’re doing research and education on artificial intelligence and data science at the intersection of mathematics and computer science. At the same time we are fully aware of the societal responsibility we have. In order to address the ethical aspects of AI and automated decision making, we have for instance included this as an important topic in the mandatory course on Philosophy and AI.”

Responsibil-IT!

“We play a significant role in developing new AI technologies like face or emotion recognition, which can be very useful in areas such as health care or security. But unfortunately it could also be used for more sinister ends.” Thuijsman cautions against throwing the baby out with the bathwater: research must continue. “Regulation is always playing catch-up with technology but you can already see e.g. many initiatives by researchers against autonomous lethal weapons.” He reckons that, like with nuclear technology, there will eventually be agencies that regulate AI developments. Weiss and Thuijsman think the benefits far outweigh the danger.

What about Berners-Lee though? Will he succeed in re-democratizing the internet? “It’s too early to say really: at the moment, Solid isn’t for end users but for developers. A lot will depend on the usability and design,” says Thuijsman. The key question is whether people are willing to change their habits and make compromises with regards to convenience. “People would have to know a bit more about data science to understand the benefits of switching. I guess the willingness will increase with every new scandal,” laughs Weiss. “But Solid has already managed to accelerate the discussion.”

By: Florian Raith (text), Paul van der Veer (photography)