Weekly Roundup 10/28/2020

Facing allegations of harming consumers by dominating the market in internet search and advertising, Google is facing an antitrust lawsuit from the U.S. Department of Justice. Similar to past antitrust cases made against AT&T and Microsoft, Google’s dominant share of the search engine market is under scrutiny. The question is whether Google’s dominance stems from producing the best product, or if the company practiced exclusionary measures to steer consumers away from its rivals.

Even if the government can prove that Google is a monopoly engaged in anticompetitive practices, they will most likely face further difficulties in determining how the search engine and digital advertising market can become more open to competitors. The suit will most likely take years to resolve and comes amidst a broader discussion on how a handful of tech companies hold immense influence and control over the tech sector’s capacity for technological innovation.

Top weekly tech headlines curated for you:

Civic Tech:

• CNAS has released its full report "Common Code: An Alliance Framework for Democratic Technology Policy." The authors call for multilateral coordination of international technology policy among democratic states to protect from authoritarian influences. The paper outlines thirteen recommendations for the world's leading "techno-democracies" on codifying international norms around data use and protection, protecting critical digital infrastructure, and investing in research and development of emerging technology.
• A number of civil rights groups have published a statement demanding that lawmakers prohibit workplace surveillance tactics used by Amazon to monitor and retaliate against workers. Amazon's "time off task" surveillance tracks employee's physical movements as they take breaks or move from one task to another. Some have expressed concerns that the TOT harvests data that prevents Amazon workers from organizing.

Open Internet:

• The continued existence of web browsers that allow users within China to access the global internet (or to "hop the Great Firewall") suggests that China could be entering a new era of openness or trying out new forms of control. Leitu, one of a number of "cross-border browsers," allows users to see content on blocked websites like Facebook and YouTube from within China legally, while still censoring political content and allowing authorities to tightly control what information users see. Use of such browsers must be tied to a phone number registered with a national ID number, so any activity on the browser can easily be connected to real-world identities. 
• The LA Times reported that Facebook has complied with requests from government officials in Vietnam to silence dissenters online, despite the company touting its commitment to free speech. Facebook suspended the account of a vocal critic of the Vietnamese government for three months before issuing a permanent ban, after the user posted criticism of land seizure in a village outside Hanoi. Vietnam's information minister reported last month that Facebook has increased its compliance with government requests for takedowns to 95%, its highest level since the platform began operating in the country in 2008.
• As mass protests against police brutality in Nigeria continue, many are concerned that the government may use internet shutdown tactics to prevent protesters from coordinating locations, crowdfunding, and receiving media coverage. A TCInsights article estimates that Nigeria could lose almost one billion dollars each week in the event of a total internet shutdown, making the economic case for maintaining an open internet.

Disinformation:

• DW Akademie has produced a number of resources on media literacy and countering misinformation to celebrate Global Media and Information Literacy Week 2020. 
• Demos, in conjunction with NDI, has just released a report that details how gendered disinformation can exclude women from public life and threatens their equal political participation when spread by state or non-state actors. The report analyzed Twitter data from Poland and the Philippines and found that actors frequently utilized highly emotive content and gender stereotypes to spread misinformation and undermine women involved in politics.
• In the most recent episode of their Diplomatic Immunity podcast, the Institute for the Study of Diplomacy interviewed NDI alumna Nina Jankowicz about her recent book on Russian information operations in Eastern Europe.
• In an opinion column for the Washington Post, three professors from the University of Virginia share that the longer users spend on Facebook, the more polarized their online news consumption becomes. In comparing the news content recommended by algorithms from Reddit and Facebook, researchers saw that Reddit's recommendations were consistently more moderate news sources than Facebook's. While Reddit gives users recommendations based on what they found to be informative or interesting (through a system of upvotes), Facebook's prioritization of "engagement" metrics leads to intense partisan content overpowering moderate voices.
• The U.S. has accused Iran of targeting voters with fake but intimidating emails after registered voters in multiple states reported receiving messages from the Proud Boys threatening to come after them unless they voted to re-elect President Trump. 

Bots and Chatbots:

• Buzzfeed News reported that a bot on the popular messaging app Telegram utilized deepfake technology to create photorealistic nude images of more than 680,000 women, whose photos were uploaded by users without consent. Amsterdam-based threat intelligence firm Sensity identified a network of over one hundred thousand users who used the free Telegram bot to produce images, often from women's public social media profile photos. The discovery has called greater attention to the threats of deepfakes for women, who could be blackmailed or extorted with these images.

Cybersecurity:

• Video conferencing platform Zoom has begun the rollout of end-to-end encryption for video calls, the first of four phases of E2EE features it plans to integrate for its users. The second phase is scheduled for rollout in 2021.
• The U.S. joined six other countries in a statement about the dangers of encryption and made new calls for backdoor encryption access by law enforcement agencies.

AI:

• Three researchers from the Institute for Defense Analyses write that the use of artificial intelligence software tools could have revealed patterns in the tactics of Russia's Internet Research Agency over time long before the 2016 election. In an article for War on the Rocks, they argue for the use of AI to identify and counteract foreign disinformation operations in the future.
• In a recent interview for the MIT Technology Review, William Isaac of Deepmind AI discusses how AI algorithms are already causing harms, but how with more investment in understanding their social biases, they could become a technological force for equality in the future.
• Access Now writes that the EU needs to do more than promote ethics in its artificial intelligence policies in the face of the risks AI has for discrimination and mass surveillance. The organization calls for mandatory impact assessments for all uses of AI.