Menu Close

Cyber Newsletter (#3) – November 2023

The European Parliament votes in favour of stricter controls on online political advertising


The European Parliament has passed new rules to regulate online political advertising as part of its new regulation on the transparency and targeting of political advertising. New rules require such advertising to be clearly identified on social networks. Information on their financing has to be made available, while the creation of a public directory is underway. Third countries will be prohibited from financing political advertising in the EU three months prior to an election or referendum. Lastly, only data explicitly provided for online political advertising purposes will be allowed to be used, excluding traditional profiling data such as religion or sexual orientation. The aim is to prevent foreign interference in European democratic processes.

​​To comply with GDPR, Meta launches subscriptions to opt out of personalized advertising on its platforms 


A few days after being called to order by the European Data Protection Board, Meta has launched subscription offers on Facebook and Instagram. Previously, the group did not explicitly ask for users’ consent to process their personal data for the purposes of advertising, a practice deemed contrary to the GDPR. Meta has now found a way to be consistent with the regulation, quickly rolling out a subscription option. In the event of refusal, users must explicitly consent to the processing of their personal data for targeted advertising purposes. Otherwise, access to social networks is denied. It remains to be seen whether this pay or opt-in option will satisfy the European authorities, although doubts remain as it is more akin to coerced consent than to voluntary and informed choice.

Nvidia unveiled a new major AI-Chip


On November 13, Nvidia announced the H200, its latest and most powerful graphics processing unit (GPU) to power artificial intelligence systems. This new product will be available starting from the second quarter of 2024. The main difference is the swap out of HBM3 for HBM3E memory. This will increase the H200’s memory capacity from 80GB to 141GB, and memory bandwidth from 3.35TB/second to what NVIDIA is expecting to be 4.8TB/second – an approximately 43% increase in bandwidth.

Japan aims to become an international leader in the chip industry


Japan’s government will allocate roughly $13 billion to support efforts to boost its chip industry. The country aims to attract new investors in the manufacturing of cutting-edge chips within the Japanese soil, in Hokkaido. In the meantime, the Japanese Prime Minister announced a new $1,2 billion subsidy to the American group Micron for the development and production of new generation electronic chips in Japan. As tensions between China and Taïwan are rising concerns about the security of the electronic chips supply, Japan seeks to become an international leader in the industry by collaborating with major international groups. 

China’s threat to Taiwan satellites network


China poses a significant cyber threat to Taiwan’s space-based infrastructure. China has developed sophisticated cyber weapons that can disrupt or destroy satellites, and it has also been known to target Taiwan’s undersea cables, which are critical for internet connectivity. Simultaneously,Taiwan is working to secure its communication satellites by developing its own low-Earth orbit (LEO) satellite communications project. It is also strengthening its cyber defense capabilities by collaborating with the U.S., concerned about the possibility of a Chinese cyberattack against their infrastructures in the event of an invasion of Taiwan. Jen Easterly, the director of the U.S. Cybersecurity and Infrastructure Security Agency, has suggested that an invasion could be accompanied by multiple cyber incidents, including sabotaging gas pipelines, contaminating water systems, seizing control of telecommunications networks, and disrupting transportation hubs. 

“Spamouflage” network: world’s largest known disinformation operation against US citizens


recent CNN report details a massive disinformation campaign originating from China, targeting American citizens. This campaign is believed to be part of a broader billion-dollar effort to shape the global information environment. The operation, believed to be the largest of its kind, employs a network of fake social media accounts and websites to spread misinformation and sow discord among the American public. These accounts post abusive messages and threats, often targeting journalists and activists. While the Chinese government has denied any involvement in the harassment campaign, experts believe that the network is linked to Chinese law enforcement. The CNN report details how the Chinese government has been using this operation to manipulate public opinion and undermine American democracy.

UNESCO presents an action plan to counter online disinformation 


UNESCO has unveiled an action plan that was eighteen months in the making, and required over 10,000 contributions from 134 countries. Sounding the alarm on the intensification of disinformation and hate speech online, the action plan presents seven fundamental principles to be respected when regulating online content. These principles include establishing independent regulators, ensuring transparency of algorithms, and educating users about critical thinking. UNESCO also addresses the importance of protecting freedom of expression, and the necessity of protecting electoral integrity during elections. 

black android smartphone on orange surface
Photo by Brian McGowan on Unsplash

French digital watchdog reports Azeri connection to disinformation campaign against 2024 Olympics.


Viginum, the French state service for Vigilance and Protection against Foreign Digital Interference, has issued a new investigation report. While Viginum found no direct connections to the Azeri government, at least one foreign actor connected to Azerbaijan has participated in a disinformation campaign aimed at discrediting France and its ability to host the 2024 Olympics. The campaign, identified by the hashtag #boycottparis2024, employed videos and pictures depicting clashes between police and protesters during the riots that followed the death of Nahel Merzouk on June 27th. The strained relationship between Azerbaijan and France has worsened since the former took control of Nagorno-Karabakh in September 2023. 

OpenAI’s turbulent November: firings, resignations and rehirings.  


This November, OpenAI found itself embroiled in a series of tumultuous events surrounding the firing and subsequent rehiring of its CEO, Sam Altman. The abrupt dismissal, stemming from concerns over transparency and trust, triggered a chain reaction of resignations, including Greg Brockman, OpenAI’s co-founder and president, and three senior researchers. Amidst the chaos, Microsoft stepped in, offering Altman and Brockman positions in a new advanced AI research team. However, after intense negotiations and internal turmoil, OpenAI reversed course, appointing a new board of directors and reinstating Altman as CEO. The rollercoaster of events left OpenAI grappling with significant disruption, low employee morale, and questions about its leadership and ability to navigate the complex ethical and commercial challenges of AI development. 

China seeks its independence in the semiconductor industry


Dutch exports to China continue to rise. In October, they increased by 29.5%as the sales of photolithography machines by ASML are the main reasons for this importation boom to China. Those machines are used to print electronic chips. New export regulations will soon be implemented in the Netherlands due to diplomatic pressure by the US. Consequently, from January 1, 2024, ASML will no longer be able to sell part of its photolithography machines to Chinese customers, without first having obtained government authorization. Therefore, the Chinese industry anticipates these changes to continue to be a powerful international actor.

The UN launched a High-Level Advisory Body on Artificial Intelligence: between support and controversies


To foster a globally inclusive approach, the UN Secretary-General is convening a multistakeholder High-level Advisory Body on AI composed of 38 experts in relevant disciplines from around the world. The Advisory Body aims to undertake analysis and advance recommendations for the international governance of AI. The first meeting of the Body took place on 27 October 2023. The initiative is still at its beginning stages, but concerns are already arising in the AI community about a potential multilateral shift in the governance of AI and relative emerging technologies.