United Kingdom
BBC accuses Perplexity AI of copyright infringement
- In a letter addressed to Perplexity’s CEO, BBC has threatened legal action for using its content ‘verbatim’ and without permission.
- BBC has not only accused them of copyright infringement but also inaccurately summarising news stories from BBC and other newspapers, thereby damaging the BBC’s reputation when Perplexity lists it as a source.
- Furthermore, the letter highlighted how Perplexity has ignored the use of robot.txt code on the BBC’s platform which is specifically designed to prevent AI tools scrapping content.
- Perplexity has refuted these claims stating that it accurately references the BBC and also provides users with the links to such articles unlike other AI models of OpenAI, Google or Meta.
- In most cases these AI models still have to comply some aspects of the UK GDPR such as showing legitimate interest for using the data, respecting robot.txt codes, referencing the sources and so on. However, copyright infringement and AI is becoming an increasingly complicated topic, which requires the development of new legislation.
A new policy seeks the removal of messaging apps from Scottish Government devices
- With a new policy coming into effect, messaging apps like WhatsApp are being removed from the Scottish Government devices.
- This follows from the publication of an externally led review by Emma Martins in December 2024- a former Channel Islands data protection commissioner.
- This policy would apply to all Scottish Government employees which prohibits unofficial communication on Government devices.
- The rationale behind this policy is to allow the Scottish Government to deliver on their openness and transparency commitments while at the same time ensuring secure, searchable data that complies with record management rules.
United States
- The New York Child Data Protection Act, which prohibits all online sites from processing personal data of anyone under the age of 18 for advertising purposes (unless strictly necessary or through consent) has finally taken effect.
- The Office of the New York Attorney General now has the power to enforce this law and seek damages or civil penalties worth up to $5000 per violation.
- This Act comes after numerous concerns for children’s mental health as prolonged social media use has been linked to increased rates of anxiety, depression and suicidal ideation among the youth. Moreover, research has found that social media companies utilise certain algorithms to keep children online for longer to as use their data for advertising purposes.
Reddit Sues Anthropic Over Unauthorised AI Training
- Anthropic which created Claude AI is being sued by Reddit for using its data to train the AI model. This comes after Anthropic issued a public statement in July 2024 stating that it had stopped the use of content from Reddit.
- However, Reddit accused Anthropic bots of accessing its website nearly 10,000 times after the statement.
- Reddit had previously offered Anthropic an opportunity to license its content much like its agreements with OpenAI and Google, but this was declined.
- These accusations specifically affect Anthropic’s claims of being an “ethics first” which has been called as “the most aggressive scraper on the internet” using other sites like YouTube to train its model.
Europe
Synchronoss Technologies Receives EU-U.S. Data Privacy Framework Certification
- Synchronoss Technologies Inc.- a leader in personal cloud platforms for telecoms- received an EU-US Data Privacy Framework certification this week.
- The DPF certification acknowledges US based organisations to be compliant with EU privacy laws such as the General Data Protection Regulation (GDPR) thereby enabling it to receive and process personal data of EU citizens.
- As a white label cloud provider, a DPF certification, reinforces Synchronoss’ commitment to international privacy standards and strengthens its position as a Tier 1 telecom operator.
- This would therefore lead to an increase in Synchronoss’ operations in the EU as it can now demonstrate its commitment to undertaking cross-border transfer of personal data with integrity, transparency and accountability.
Telekom and NVIDIA announce plans to build industrial AI Infrastructure in Germany
- Nvidia and Dutch Telekom are set to build the world’s first industrial AI cloud in Germany for European manufacturers. This will not only allow Germany to establish its position as an industrial hub for AI but also pave the way for the development of large-scale facilities aimed at training next-gen AI models.
- In this partnership, NVIDIA will provide 10,000 GPUs via the DGX B200 systems and Omniverse accelerated workloads from leading software vendors. Dutch Telekom, on the other hand, will provide the infrastructure for the data centre along with handling the operations, sales, security and AI solutions.
- Jensen Huang, the CEO of Nvidia, called this AI infrastructure as key for creating the intelligence required for Europe’s ‘leading industrial companies to advance simulation-first, AI-driven manufacturing.’
International
‘Cyber Fattah’ Hactivist Group Leaks personal data from Saudi Games Website
- Cyber Fattah leaked thousands of records of visitors and athletes of Saudi Games- an annual sporting event since 2022 through Telegram in the form of SQL dumps.
- The data, most likely extracted from the Saudi Games website included information such as IT Staff credentials, government email addresses, passports, medical forms, bank statements and other sensitive documents.
- This attack highlighted the lack of security within the platform which stores such sensitive data.
- Cyber Fattah in collaboration with other hactivists such as 313 team, LulzSec Black and Cyber Islamic Resistance has increased activities targeting data in the US, Israel and other NATO countries.
- Overall, events such as large-scale sporting games may become an increasingly attractive target for such malicious actors due to the financial, political and strategic incentives.
