Tech Brief - Last EU-US tech council before elections, cloud sovereignty requirements deleted from EUCS

“Regardless of whoever is going to be next year in the White House or in the Berlaymont, this strength has to continue to be nurtured and developed.”

– An EU official said at the EU-US Trade and Technology Council in Leuven, Belgium, on Thursday.

Story of the week: The European Commission played down fears that the EU-US Trade and Technology Council (TTC) could succumb to a Donald Trump presidency as pressure on high-level officials from both sides increased ahead of their meeting in Leuven, Belgium, on Thursday. “We believe that the EU-US transatlantic relationship is there to stay and has a lot of value added on both sides,” an EU official said. “The Commission was setting up “a structured internal process to prepare for all possible outcomes from the United States presidential elections,” the official added. However, with US polls showing Trump leading in six of the seven swing states ahead of November’s presidential elections, European stakeholders have voiced concerns that this week’s forum – the sixth since its inception – could be the last of its kind, adding uncertainty to the pre-existing scepticism around the TTC’s effectiveness. Read more.

Don’t miss: French MPs and EU cybersecurity agency ENISA are due to make landmark decisions on cloud cybersecurity on 10 and 15 April, amid concerns of espionage and fragmentation of the European cloud market. French MPs will vote on an all-encompassing digital bill (SREN), including specific cloud sovereignty provisions, finishing a year-long legislative process. Shortly afterwards, ENISA’s dedicated working group (ECCG) will meet to approve the EU cybersecurity certification scheme (EUCS), which aims to harmonise national cloud certifications. In both documents, cloud sovereignty requirements have been contentious. Read more.

Also this week:

Before we start: If you’re looking for more tech analysis, tune in to our weekly podcast.

Today's edition is powered by VivaTech

EU-US cooperation. The EU and US are in close contact over artificial intelligence (AI) risks and mitigation, including a possible partnership for a framework on generative AI, according to an unreleased draft statement, dated 13 March and seen by Euractiv last Friday, for the EU-US Trade and Technology Council held on 3-4 April. The draft statement was tabled for discussion on the EU side at last week’s meeting of EU ambassadors. Read more.

Romania to tackle spread of deepfakes during elections. The Romanian government is set to approve a memorandum to tackle the spread of deepfake content during election campaigns, Research Minister Bogdan Ivan announced on Thursday. Meanwhile, the Romanian government, major technology companies, the Central Electoral Board, and seven other Romanian institutions are in talks to create an ‘Election Hub’, which will establish a direct channel between the Romanian authorities and major social media platforms to ensure that strict measures are in place to prevent the spread of deepfake content. Read more.

US-UK partnership. The United States and Britain formed a new partnership focused on the safety of AI on Tuesday, aiming to address concerns about upcoming advanced AI technologies. US Commerce Secretary Gina Raimondo and British Technology Secretary Michelle Donelan signed a memorandum of understanding to jointly work on developing advanced AI model testing, building on commitments made at the AI Safety Summit in Bletchley Park in November.

NGOs on AI Act. NGOs, including EDRi, AccessNow, ARTICLE19, Amnesty International, and others, published their analysis of the EU’s AI Act on Wednesday, writing that while the regulation shows progress in accessibility and transparency, concerns remain regarding exemptions for law enforcement, national security, and partial bans on harmful surveillance practices. Civil society’s involvement is crucial for effective implementation, and global lawmakers should heed lessons from the EU’s experience to ensure robust AI regulations, the document notes. Another NGO, the Civil Liberties Union for Europe, also published its analysis of the AI Act on Thursday, highlighting that it “fails to effectively protect the rule of law and civic space.”

New supercomputer. Microsoft and OpenAI are collaborating on a data centre project potentially costing $100 billion. The project includes plans for an artificial intelligence supercomputer named “Stargate,” slated to launch in 2028, The Information reported last Friday.

No need to sign up anymore. OpenAI, backed by Microsoft, will make its ChatGPT generative AI chatbot accessible without requiring users to sign up, it was announced on Monday. The San Francisco-based company will implement additional content safeguards for users accessing ChatGPT without signing up, including blocking prompts and generations in unspecified categories.

EU, US to share data on semiconductor market distortion. The EU and US will share results from respective market surveys on possible market distortions in the semiconductor industry, officials from the two allies said in a press conference at the sixth meeting of the EU-US Trade and Technology Council (TTC) on Friday. Read more.

Teams and Office unbundling. Microsoft will sell its Teams chat and video app separately from its Office suite globally, the tech giant told Reuters on Monday. The decision comes after complaints from rivals and feedback from the European Commission. Separating the products aims to provide clarity for customers and flexibility for multinational companies in their purchasing decisions. Read more.

Olympic Games Russian cyberattacks. French President Emmanuel Macron said on Thursday that he had no doubt Russia would malevolently target the Paris Olympics this summer, in comments that underline the fraught geopolitical backdrop to the games. Read more.

AT&T data breach. AT&T announced last Saturday that it is investigating a data set released on the dark web Reuters reported on Monday. Preliminary analysis suggests it affects approximately 7.6 million current account holders and 65.4 million former account holders. The data appears to be from 2019 or earlier, and contains sensitive information such as names, addresses, birth dates, and Social Security numbers — but not personal financial information or call history, according to AT&T. The company stated there is no evidence of unauthorised access to its systems resulting from this incident and it is not yet known whether the data originated from AT&T or from one of its vendors.

Google deleting data. Google has reached a proposed legal settlement in a class action lawsuit filed in the US in 2020, accusing the company of invading users’ privacy by collecting data even in “private mode,” BBC reported on Monday. The settlement involves deleting billions of records and accepting restrictions on user tracking. While Google supports the deal, it disputes the accusations and has already implemented changes in response to the lawsuit.

Meanwhile in the US: women’s faces stolen. A Washington Post article published last Thursday reported on a controversial ad campaign that used AI-generated images of women without their consent in the US. There is currently no federal law specifically addressing deepfakes, and although more than three dozen state legislatures are working on AI-related bills, regulations concerning deepfakes are primarily focused on political ads and nonconsensual pornography.

Also in the UK. A BBC article published on Tuesday recounts the experience of “Jodie”, who discovered that images of her were used in deepfake pornography, only to realise that her best friend was behind it. It mentions the criminalisation of sharing deepfake porn under the UK’s Online Safety Bill but emphasises Jodie’s frustration that the law does not penalise those who commission or create deepfakes. The situation is not much better in the EU either, as last October Euractiv also reported that nude deepfakes, including those of minors, are becoming increasingly common online as the tools to create them become more accessible – yet the law is still behind in regulating such material.

Reform WhatsApp. The Mozilla Foundation is calling on WhatsApp to implement changes to combat election disinformation ahead of major elections in various countries. With over 2 billion users, WhatsApp is a major platform for spreading political disinformation, which includes deepfakes, inaccurate reporting, hate speech, and conspiracy theories, Mozilla pointed out in a post and also in a letter to WhatsApp, on Tuesday. The foundation’s research highlights WhatsApp’s lack of intervention compared to other social media platforms like the also Meta-owned Facebook and Instagram.

Greece sees its biggest GDPR fine. The Greek data protection authority issued its biggest fine under the GDPR. The agency fined the Ministry of Immigration and Asylum €175,000 for its use of software, including facial recognition, to target migrants and NGO workers.

Vestager in the Parliament. On Wednesday, the Executive Vice President of the European Commission Margrethe Vestager attended an exchange of views in the European Parliament’s Committee on Internal Market and Consumer Protection, where MEPs could ask her about the compliance and implementation of the Digital Markets Act (DMA). Vestager talked about, among other things, her hopes for a global effect of the DMA, and noted that more tools will likely be needed in the future as technology develops.

You are late. Internal Market Commissioner Thierry Breton, has sent reminders to 21 member states urging them to designate national Digital Service Coordinators (DSCs) for the Digital Services Act (DSA), Contexte reported on Tuesday, emphasising the importance of a strong network, Breton stressed that the network’s effectiveness depends on every member state fulfilling its obligations. The deadline for each member to designate its DSC, to help enforce the regulation, was 17 February, but not all of them fulfilled the requirement.

BlaBlaCar’s boost. On Wednesday, the French car-sharing application BlaBlaCar announced it had secured €100 million credit facility in order to grow. It aims to follow its strategy combining several modes of shared transportation, through its bus activities in Europe and Brazil. BlaBlaCar has been financially profitable for two years and claims to have saved 2 million tonnes of CO2 in 2023.

EU-US critical raw material partnership. The EU and US are enticing developing countries into a partnership that looks to align measures on the supply of critical raw materials, a European Commission official said on Wednesday. Roughly 24 countries have been invited to attend a launch event of the Minerals Security Partnership (MSP) Forum on Friday, during the EU-US Trade and Technology Council, said the official. “What was missing in the MSP was outreach to resource-rich developing countries,” the official said. The EU and US have “a new and potentially better offer [for the developing nations] that creates more value added in those countries,” they said. Read more.

Microsoft announces breakthrough in quantum. Microsoft and Quantinuum announced on Wednesday that they had managed to demonstrate the “most reliable qubits on record,” claiming the error rate has been improved by 800 times. Quantinuum and Microsoft managed to create four logical qubits, which they claim is “a historic moment”. According to experts, 100 logicial qubits will be required to build a quantum computer with computing capacities equivalent to traditional computers.

Chinese chips. China is reportedly developing techniques to produce advanced chips without relying on the Dutch semiconductor giant ASML’s lithography machines. China aims to make 5-nanometer grade chips, with Naura Technology Group and Huawei Technologies involved in this effort, potentially challenging US restrictions on China’s chip-making capabilities.

ECLAG disagrees. A document dated last Friday, seen by Euractiv, highlights feedback from the European Child Sexual Abuse Legislation Advocacy Group (ECLAG) on the Belgian Presidency’s latest proposal to address child sexual abuse online. ECLAG urges revisions to proposed approaches, emphasising the need to protect children’s rights and strengthen detection measures across all online services. They stress the importance of including survivors in relevant boards and detecting grooming in all online services.

“A clumsiness”. The president of the French National Assembly’s commission of inquiry on terrestrial television, Quentin Bataillon (Renaissance, Renew), stirred a controversy with his comments on Tuesday. Although the commission’s inquiry work is still ongoing, and expected to deliver its recommendations in April or early May, Bataillon went on air. He publicly criticised famous TV journalist Yann Barthès, in an emission presented by Cyril Hanouna, a competitor of Barthès. Bataillon equally slammed the commission’s rapporteur Aurélien Saintoul (LFI, The Left). Several politicians from different left-wing parties publicly criticised Bataillon and demanded his resignation. Renew members, including the president of the National Assembly, Yaël Braun-Pivet, and the president of the Renew group at the National Assembly, Sylvain Maillard, contacted Bataillon, who recognised “a clumsiness” he said he regretted on Thursday.

Elan: new media dedicated to entering the labour market. Hugo Travers, 26, who created the successful media HugoDécrypte, announced last Sunday the launch of a new media, Elan, dedicated to young people entering the labour market. Elan will offer free videos on Instagram, TikTok, YouTube, and LinkedIn and a recruitment platform. Founded in 2015, HugoDécrypte has more than 14 million followers on its different platforms. Elan’s mission is to become a platform where young people discover new industries, jobs, careers, while providing guidance on housing, financing, health, the administration in order to “know everyone’s rights” or “advice on work-life balance”. On Tuesday, Hugo Travers claimed that one million people already followed Elan’s content.

Possible French digital addiction regulation. A debate over children’s addiction to digital technology, and what can be done about it, is gaining traction in France, while at the EU level, there are rumblings for regulation on addictive design. French delegated minister for youth Sarah El Haïry and digital secretary of state Marina Ferrari launched a new initiative on Thursday called ‘P@rents, let’s talk about digital’. Announced in the newspaper Le Parisien by El Haïry, the initiative will set up free workshops countrywide for parents, discussing all topics related to the use of digital technologies by children. The workshop initiative follows an earlier promise by President Emmanuel Macron. Read more.

French Senators blame Macron government for SREN delay. On Tuesday, French Senators blamed the Macron government for the eight-month delay in the legislative process on a landmark digital bill, “sécuriser et réguler l’espace numérique”, shortened SREN, saying that the government had failed to prepare the text effectively. The European Commission objected to some provisions in the legislation, which delayed the process in France. Opposition politicians said the Commission’s objections were the result of bad preparation on the French government’s part. Read more.

Artists not satisfied. In a letter published on Monday, 200 artists urged tech platforms to halt the devaluation of music, arguing that streaming services and social media platforms exploit artists, leading to inadequate compensation and undermining music’s value. They call for fair compensation, transparency in streaming royalties, and support for independent artists.

OpenAI to breach YouTube rules? Although he lacked direct knowledge of whether OpenAI had indeed employed YouTube videos for this purpose, Neal Mohan, CEO of YouTube, said that using YouTube videos to train OpenAI’s text-to-video generator, Sora, would violate the platform’s terms of service, Bloomberg reported on Thursday.

French Excel sheet competitor becomes a unicorn. Pigment, a French startup selling a SaaS solution to improve corporate decision making, became the country’s latest unicorn. Its value exceeded $1 billion since it raised an additional $145 million. The startup presents its solution as an alternative to Microsoft Excel sheets. Economy Minister Bruno Le Maire congratulated the founders on LinkedIn, saying that “this new fundraising proves that France has incredible talents”.

Commission vetoes Maltese broadband law. On Tuesday, the European Commission announced it had vetoed a Maltese regulation by the Malta Communications Authority (MCA) of the wholesale broadband market. The MCA regulation aimed at regulating the wholesale physical and virtual infrastructure access. The Commission raised concerns at the end of January, while the body of European regulators issued an opinion in February disagreeing with the Commission. After a four-month analysis, the Commission decided to nevertheless veto the regulation, considering that the MCA’s analyses were not sound. It wrote that “the MCA’s justification for identifying GO, one of the two nationwide networks, as having significant market power at the wholesale level was found to be insufficient”. It furthermore reminded that “the Commission’s veto decision highlights the importance of a competitive analysis that considers all market dynamics but does not prejudge the presence of a competition problem at the retail level, which still should be proven by the regulator.”

EU officials scramble to prove transatlantic forum is useful. The European Commission played down fears that the EU-US Trade and Technology Council (TTC) could succumb to a Donald Trump presidency as pressure on high-level officials from both sides increased ahead of their meeting in Leuven, Belgium, on Thursday. Read more

[Edited by Zoran Radosavljevic]

Thanks for reading. Be sure to spread the word and come and say hello on X.