Once again, artificial intelligence took center stage in discussions at the yearly Web Summit held in Lisbon, Portugal. On the concluding day of the event, whistleblower and activist Chelsea Manning voiced apprehensions regarding the advancement of artificial intelligence and the growing conflict between humans and technology.
“We live in this era where we’re surrounded by technology […] I’ve encountered numerous instances of this clash between the humans that we are and the tech, and the way technology is shaping how we interact with each other, and how institutions are a part of that,” said Manning.
Manning asserts that her experiences, from homelessness to military service and imprisonment for divulging military documents to WikiLeaks in 2010, have been profoundly influenced by her engagement with the digital realm. “I let that drive how I integrate, view technology, and want to build a bigger, better future,” Manning continued.
Manning also raised questions about the intent behind the swift progress of AI technology, pondering who stands to ultimately benefit from it and whether the expansion into artificial intelligence is transforming everyday life into a commodity system.
“Are we simply forgetting the human element and turning every single interaction that we have on a day-to-day basis into a commodity?” Manning said. “This commodity is us doing something as simple as swiping on an app or clicking a cute little heart icon.”
Referring to artificial intelligence and large language models as the trending buzzwords, Manning highlighted the substantial volume of data used to train these models, providing examples such as Wikipedia, textbooks, videos, and images.
She further pointed out that unlike files stored on a server, this data is ingrained in the model’s weights and balances, making it difficult to erase. This contrasts with the European Union’s General Data Protection Regulation (GDPR) and privacy rights, which allow individuals to request the deletion of their data.
“I think that the onus is on us, as technologists, to develop tools that take into account the fact that people need to have some semblance of privacy and that communications shouldn’t be a resource to be extracted and to be commodified,” Manning said. “I don’t want to live in a world where the relationships that we have are a numerical value.”
Manning issued a warning about large language models, emphasizing that once an individual’s personal or private information is assimilated, it becomes an indelible component of the model, rendering removal unattainable.
Manning also touched on the issue of the surge and escalation of AI-generated deep fakes. “The things that have caught my attention in the last six months have been not the fact that people are using deepfakes in public discourse and discussion, but the fact that they can be made and the fact that they exist, and are used to cast doubt on actual verified information […] I think that this trend keeps causing people to question their own reality more and more.”
With the rapid emergence of deep fakes, policymakers are concerned about its potential impact on upcoming elections and its contribution to online hate. Manning suggested a solution involving the use of cryptography to authenticate information.
In 2021, Manning assumed the role of a security consultant and privacy advocate at Nym Technologies, a Switzerland-based company. Nym Technologies introduced its privacy-centric project, Nym, a decentralized identity platform, on the Cosmos blockchain in 2020.
“I think that my focus more broadly and generally is on the importance of having some semblance of privacy—through encryption or through the technical, technological means that we have—and be able to use that intuitively […] It shouldn’t be hard, and it shouldn’t be difficult to use.”
