Data and security

poland-investigates-chatgpt-after-alleged-data-privacy-breach

Poland investigates ChatGPT after alleged data privacy breach

Data and security, Deep tech, Government and policy / /

Poland’s data protection watchdog is investigating OpenAI’s ChatGPT after an unnamed complainant accused the company of GDPR breaches.

“The case concerns the violation of many provisions on the protection of personal data, which is why we will ask OpenAI to answer a number of questions in order to thoroughly conduct the administrative proceedings,” said Jan Nowak, president of the country’s Personal Data Protection Office (UODO).

He added that “these aren’t the first doubts” about the AI tool’s compliance with European principles of data privacy and security.

According to the UODO, the accusations are structured around “unlawful and unreliable” data processing and lack of transparency.

The <3 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol’ founder Boris, and some questionable AI art. It’s free, every week, in your inbox. Sign up now!

Specifically, the complainant alleged that ChatGPT generated false information about them and did not proceed to the required correction following a formal request. They also claimed that they were unable to find which of their personal data was processed by the company, and received “evasive, misleading, and internally contradictory” answers to questions.

In response, the Polish watchdog will examine the case and clarify any doubts surrounding OpenAI’s systemic approach to European personal data protection rules. Nevertheless, the proceedings “will be difficult” as the company is located outside of the Union, said the regulator.

“The development of new technologies must respect the rights of individuals arising from, among others, the GDPR,” noted the agency’s deputy president Jakub Groszkowski. “The task of the European personal data protection authorities is to protect European Union citizens against the negative effects of information processing technologies.”

This isn’t the first time OpenAI has faced scrutiny in the bloc.

Last March, Italy became the first Western country to impose a temporary ban on ChatGPT, after its data protection agency (again) accused the company of “unlawful” collection of personal data and the absence of an age verification system for minors. In the same month, the European Consumer Organisation (BEUC) called for EU and national authorities to investigate OpenAI’s system.

Published

Back to top

Poland investigates ChatGPT after alleged data privacy breach Read More »

uk-targets-meta’s-encryption-plans-over-child-sexual-abuse-concerns

UK targets Meta’s encryption plans over child sexual abuse concerns

Data and security, Government and policy / /

On Wednesday, the UK’s Home Secretary Suella Braverman unveiled a new campaign against Meta, urging the tech giant to rethink its plan to roll out end-to-end encryption (E2EE) on Facebook Messenger and Instagram.

The company aims to finalise the encryption rollout later this year, but the British government is worried that the move will hinder the detection of child sexual abuse.

According to the Home Office, 800 predators are currently arrested per month and up to 1,200 children are protected from sexual abuse following the information provided by social media companies. If Meta’s encryption moves forward, the National Crime Agency (NCA) estimated that 92% of Messenger and 85% Instagram direct referrals could be lost.

Based on these risks, Braverman is asking Meta to implement “robust safety measures” that ensure minor protection, or halt the encryption rollout altogether.

“The use of strong encryption for online users remains a vital part of our digital world and I support it, so does the government, but it cannot come at a cost to our children’s safety,” Braveman said in a statement.

The Home Secretary first outlined her concerns in a letter to Meta thispast July. But the company “has failed to provide assurances” ensuring protection from “sickening abusers,” she now noted, adding that “appropriate safeguards” are an essential requirement for its end-to-end encryptionplans.

In response (or even anticipation) of the government’s attack, Meta published yesterday an updated report on its safety policy for the messaging platforms.

“We are committed to our continued engagement with law enforcement and online safety, digital security, and human rights experts to keep people safe,” the company writes in the report — which includes measures such as restricting adults from messaging teens they’re not connected to on Messenger.

Nevertheless, the tech giant stresses its commitment to delivering end-to-end encryption as standard for Messenger and Instagram.

“We strongly believe that E2EE is critical to protecting people’s security. Breaking the promise of E2EE — whether through backdoors or scanning of messages without the user’s consent and control — directly impacts user safety,” argues the report.

But the UK government may be holding the upper hand in the dispute, now armed with the Online Safety Bill, passed by the parliament on Tuesday.

The legislation sets sweeping content rules for social media, even empowering Britain’s comms regulator, Ofcom, to force tech companies to monitor messaging services for child sexual abuse content — a provision that has, reasonably, sparked controversy.

Published

Back to top

UK targets Meta’s encryption plans over child sexual abuse concerns Read More »

hackers-breach-defences-of-icc-war-crimes-tribunal

Hackers breach defences of ICC war crimes tribunal

Data and security, Government and policy / /

The computer systems of the International Criminal Court (ICC), the world’s most high-profile war tribunal, were hacked last week, according to a statement released by the court yesterday.

The tribunal said that its services detected “anomalous activity” affecting its information systems and that “immediate measures” were adopted to respond to this cybersecurity incident and to mitigate its impact. 

The ICC, headquartered in The Hague, Netherlands, is the only permanent war crimes tribunal and handles extremely sensitive data about some of the world’s most horrific atrocities. 

At this time, the nature of last week’s cybersecurity incident remains unclear and it’s not yet known whether any data held on the ICC’s systems was accessed or exfiltrated.

However a source told Dutch broadcaster NOS that a large number of sensitive documents have been captured, but the ICC has not confirmed this. A spokesperson for the court did not immediately respond to our request for comment. 

The court did say however that it is currently investigating the incident in collaboration with Dutch authorities, and that it “continues to analyse and mitigate the impact of this incident.” It also added that it will build on its existing work to strengthen its cybersecurity framework, including increased adoption of cloud technology. 

Established in 2002, the court is currently investigating crimes against humanity in 17 states including Ukraine, Uganda, Venezuela, and Afghanistan. For instance, in March 2023, the ICC issued an arrest warrant for Russian President Vladimir Putin concerning crimes linked to Russia’s invasion of Ukraine.

The Dutch intelligence agency said in its 2022 annual report that the ICC was “of interest to Russia because it is investigating possible Russian war crimes in Georgia and Ukraine.”  

In August 2023, ICC Prosecutor Karim Khan warned that cyber attacks could be part of future war crimes investigations and that the ICC itself could be vulnerable and should strengthen its defences.  

“In all probability this is a nation state attack [by Russia] happening just a week after the ICC established a field office in Kyiv to track Russian war crimes,” Jamie Moles, senior technical marketing manager at US-based cybersecurity firm ExtraHop, told TNW. 

He continued: “It seems the ICC may have lost significant volumes of data in an attack, the details of which it refuses to disclose at this time. Too often we see institutions fail to properly secure their networks and data leading to breaches and stolen data. No one is exempt from bad actors, which is why every organisation should prepare to be attacked.” 

Hackers breach defences of ICC war crimes tribunal Read More »

uk’s-controversial-online-safety-bill-to-become-law

UK’s controversial Online Safety Bill to become law

Corporates and innovation, Data and security, Government and policy / /

Britain’s controversial Online Safety Bill will soon become law after passing through parliament on Tuesday.

The sweeping legislation places strict news content moderation rules on social media companies. Platforms will become legally responsible for the material they host.

Under the new rules, platforms will have to quickly remove any illegal content — or stop it from appearing in the first place. They also must prevent children from accessing harmful and age-inappropriate content, and enforce age limits and age-checking measures.

Those that fail to take rapid action face fines up to £18mn (€20.8mn) or 10% of their global annual revenue — whichever is biggest. In some cases, executives of platforms could even be imprisoned.

Michelle Donelan, the UK’s technology minister, said the rules would “make the UK the safest place in the world to be online.”

The <3 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol’ founder Boris, and some questionable AI art. It’s free, every week, in your inbox. Sign up now!

“The Online Safety Bill is a game-changing piece of legislation,” she said.

Children’s charities have welcomed the legislation, but digital rights activities and tech companies have raised alarm about certain implications.

Messaging platforms have been particularly opposed to the potential scanning of encrypted messages, while privacy advocates fear that free speech will be restricted. Wikipedia, meanwhile, has warned that it won’t comply with the requirement for age checks. The online encyclopedia has even threatened to withdraw from the UK over the rules.

In the six years since the bill was first proposed, some of the concerns have been addressed by amendments. Notably, lawmakers last year replaced the focus on “legal but harmful” content with an emphasis on child protection and illegal content. The government has also promised to protect end-to-end encryption, but critics have dismissed the pledges as “delusional.”

Donelan has sought to allay their fears.

“Our common-sense approach will deliver a better future for British people, by making sure that what is illegal offline is illegal online,” she said. “It puts protecting children first, enabling us to catch keyboard criminals and crack down on the heinous crimes they seek to commit.”

Published

Back to top

UK’s controversial Online Safety Bill to become law Read More »

85%-of-eu-citizens-want-more-political-action-against-disinformation

85% of EU citizens want more political action against disinformation

Data and security, Government and policy / /

The fight against “fake news” appears to have overwhelming support in the EU.

According to a new study, 85% of the bloc’s citizens want policymakers to take more action against disinformation, while 89% want increased efforts from platform operators. Just 7% do not feel that stronger responses are required.

The findings emerged from surveys by Bertelsmann Stiftung, a pro-business German think tank with close ties to the EU.

Across the EU, 54% of respondents said they were “often” or “very often” unsure whether information they found online was true. However, only 44% of them had recently fact-checked the content they’d seen. 

Younger and more educated people were more likely to take active response to false information. Those in favour of combating disinformation tended to be further to the political left.

Table showing responses to false information (EU-wide and by country).
Responses to false information (figures in percentages). Credit:  (EU-wide and by country). Bertelsmann Stiftung

In response to the findings, Bertelsmann Stiftung made the following recommendations:

  1. Establish an effective system for monitoring disinformation both in Germany and across Europe.
  2. Raise public awareness about the issue of disinformation.
  3. Promote media literacy among people of all age groups.
  4. Ensure consistent and transparent content creation on digital platforms. 

Such interventions, however, have proven divisive. Around the world, politicians have been accused of exploiting concerns around disinformation to suppress dissent and control narratives.

In the UK, campaigners found that government anti-fake news units have surveilled citizens, public figures, and media outlets for merely criticising state policies. The units also reportedly facilitated censorship of legal content on social media.

The EU, meanwhile, recently adopted the Digital Services Act (DSA), which requires platforms to mitigate the risks of disinformation. Opponents of the law fear that it will lead to state censorship.

Critics have also raised alarm about tech firms acting as arbiters of truth. But Bertelsmann Stiftung’s researchers argue that more intervention is essential.

“People in Europe are very uncertain about which digital content they can trust and which has been intentionally manipulated,” Kai Unzicker, the study author, said in a statement.

“Anyone who wants to protect and strengthen democracy cannot leave people to deal with disinformation on their own.”

Stronger responses could also bolster the growing flock of anti-disinformation startups.

The emerging sector is dominated by the US, but a hub is also emerging in Ukraine, where technologists are turning lessons from fighting Russian propaganda into new businesses.

Published

Back to top

85% of EU citizens want more political action against disinformation Read More »

eu-president:-europe-is-the-‘global-pioneer’-of-citizen’s-digital-rights

EU president: Europe is the ‘global pioneer’ of citizen’s digital rights

Data and security, Government and policy, Insider / /

Europeans have become “pioneers in online rights” and now want to lead a “global framework for AI,” the EU’s top official said today.

Ursula von der Leyen, the European Commission’s president, revealed the bloc’s digital plans during her State of the Union address in Strasbourg. She used the speech to flaunt the achievements of her three-year reign.

A particularly large spotlight was shone on her tech policies.

“We have set the path for the digital transition and become global pioneers in online rights,” von der Leyen said.

The former German defence minister praised the bloc’s work on semiconductor self-sufficiency, which centres on the Chips Act. Backed with €43bn of funding, the legislation aims to double the EU’s market share in semiconductors to at least 20% by 2030.

The <3 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol’ founder Boris, and some questionable AI art. It’s free, every week, in your inbox. Sign up now!

Von der Leyen also touted the union’s clean tech industry, as well as the digital projects in NextGenerationEU, a COVID-19 recovery plan. Her biggest brag, however, involved digital safety.

“Europe has led on managing the risks of the digital world,” she said.

To the chagrin of Silicon Valley, the EU has become the world’s most formidable tech regulator. Tough laws on privacy, tax avoidance, antitrust, and online content have led to eye-popping fines for some of the biggest companies in the US.  Von der Leyen warned them that more rules are coming.

To justify the intervention, she argued that disinformation, data exploitation, and “harmful content” have reduced the public’s trust and breached their rights.

“In response, Europe has become the global pioneer of citizen’s rights in the digital world,” she said.

As evidence for this claim, von der Leyen pointed to two recent regulations: the Digital Services Act (DSA), which imposes rules on content moderation, and the Digital Market Act (DMA), which aims to reign in big tech’s dominance.

Her next big target is artificial intelligence.

“We need an open dialogue with those that develop and deploy AI.

In recent months, concerns have grown about AI causing job losses, discrimination, surveillance, and even extinction. To mitigate the threats, the EU will soon adopt the AI Act, the first-ever comprehensive legislation for the tech.

Von der Leyen described the rules as “a blueprint for the whole world.” She also laid out the next steps of the EU’s plan.

“I believe Europe, together with partners, should lead the way on a new global framework for AI, built on three pillars: guardrails, governance, and guiding innovation,” she said.

The main guardrails will be provided by the AI Act. For governance, von der Leven called for the creation of a global panel of scientists, tech companies and independent experts. Together, they would inform policymakers about developments in the field.

On innovation, she announced a project that will enable AI startups to train their models on the EU’s high-performance computers. The private sector, however, will likely want further support. In an open letter published in June, some of Europe’s biggest companies warn that the AI Act will inhibit innovation and jeopardise the continent’s businesses.

Von der Leyen, meanwhile, called for closer collaboration with the private sector.

“We need an open dialogue with those that develop and deploy AI,” she said. “It happens in the United States, where seven major tech companies have already agreed to voluntary rules around safety, security and trust.

“It happens here, where we will work with AI companies, so that they voluntarily commit to the principles of the AI Act before it comes into force. Now we should bring all of this work together towards minimum global standards for safe and ethical use of AI.”

EU president: Europe is the ‘global pioneer’ of citizen’s digital rights Read More »

this-patrolling-security-robot-wants-to-guard-your-premises

This patrolling security robot wants to guard your premises

Data and security, Startups and technology / /

A Swiss startup has unveiled a solution to the global shortage of security guards: an autonomous patrol robot.

Named the Ascento Guard, the two-wheeled sentinel is equipped with thermal and infrared cameras, speakers, a microphone, and GPS tracking. The bidepal design promises all-terrain mobility, fall recovery from any position, and top speeds of 5km/h.

Using these features, the Ascento Guard can spot trespassers, monitor parking lots, and record property lights. It can also identify floods and fires, as well as check that doors and windows are closed.

When an incident is detected, an alarm is sent to an operator. Only then is a human security guard sent onsite to take action.

The <3 of EU tech

The latest rumblings from the EU tech scene, a story from our wise ol’ founder Boris, and some questionable AI art. It’s free, every week, in your inbox. Sign up now!

The bot is the brainchild of Ascento, a Zurich-based developer of bipedal security robots. Alessandro Morra, the startup’s CEO, told TNW that Ascento Guard is designed for large outdoor premises. 

“Instead of installing many fixed cameras or sending human guards in harsh weather conditions and at night for patrol, Ascento Guard can secure the assets,” he said.

The Ascento co-founders: Dominik-Mannhart, Alessandro-Morra, Ciro Salzmann, Miguel de la Iglesia Valls
Ascento co-founders (left to right) Dominik-Mannhart, Alessandro-Morra, Ciro Salzmann, and Miguel de la Iglesia Valls. Credit: Ascento

Ascento’s founding team combines experience as security guards with robotics expertise honed at ETH Zurich, a renowned research university.

Their inventions have been deployed at large outdoor warehouses, industrial manufacturing sites, and pharma campuses. Since the start of this year, the robots have secured over 3,000 km of outdoor premises.

The Ascento Guard is the latest addition to the portfolio. According to its creators, the bot can be installed and deployed within a few hours.

Just like a human security officer, the Ascento Guard can be hired by the hour. Autonomous charging will then keep the device running at speeds of up to 5km/h.

A picture of the Ascento web interface, which shows surveillance statistics including the location of any incidents
Operators can monitor the surveillance in a web interface. Credit: Ascento

A companion app extends the robot’s capabilities. The app integrates with existing video management systems, offers end-to-end encrypted two-way communication, and generates security reports.

Morra is particularly excited about the system’s AI analytics. He envisions them identifying suspicious patterns, such as specific locations and times of incidents, or cars that consistently park in distinctive places.

“This robot design is just the beginning,” Morra said. “We are seeing multiple opportunities for how we can complement our robot to offer an indoor, aerial application integration of our technology.”

The live view of the web interface, which shows footage from cameras at the left, back, and front of the premises, as well as readings from a thermal camera
The web interface also provides a live view of footage from the cameras. Credit: Ascento

Alongside the new robot launch, Ascento today announced that it’s received another $4.3mn in funding. The pre-seed round was led by VC firms Wingman Ventures and Playfair Capital.

You can review their investment for yourself by watching the video below: