The Guardian · 5 Jun, 04:49
Australian authorities have seized over 100,000 live exotic cockroaches worth $200,000 from a breeder in New South Wales, in a record-breaking bust linked to the pet trade. The cockroaches, including dubia and Madagascar hissing cockroaches, were likely destined for use as food for pet reptiles.
Biosecurity officials from the Australian federal environment department seized more than 100,000 live exotic cockroaches from a commercial breeder in Bathurst, New South Wales. The seized cockroaches included dubia cockroaches and Madagascar hissing cockroaches, one of the world's largest cockroach species. The animals have a commercial value of up to $200,000 and were likely destined for the pet trade, including use as food for pet reptiles. This seizure is the largest of illegal exotic invertebrates in Australia.
Why it matters: The seizure is significant because exotic cockroaches have not been subject to an environmental risk assessment, and their presence in Australia can spread disease and harm native wildlife and agriculture. The cockroaches cannot be legally imported, and therefore, cannot be legally kept, bred, or sold in Australia. The incident highlights the risks associated with the illegal exotic wildlife trade and the need for stricter regulations and enforcement.
100,000 (number of cockroaches seized), $200,000 (commercial value of seized cockroaches), 500 (number of individual native cockroach species in Australia)
“We take our job protecting Australia’s unique biodiversity and breaches of national environment law very seriously,” - a spokesperson for the Department of Climate Change, Energy, Environment and Water
The Guardian · 5 Jun, 06:00
New claimants are seeking to sue Elon Musk's company xAI over demeaning, sexualized material created by its Grok AI tool, following a test case launched by Labour MP Jess Asato. The claimants argue that xAI is liable for the content generated by its AI model.
Labour MP Jess Asato has launched a test case against Elon Musk's company xAI, seeking damages over the creation and circulation of fake images of her in a bikini and an AI-created video that she said showed her 'being chloroformed and prepared for a sexual assault'. Several new claimants have come forward to take legal action against xAI over similar demeaning content generated by its Grok AI tool. The claim argues that xAI violated data protection law and breached Asato's private information when it allowed the images to be generated. The legal action aims to hold xAI liable for the design choices made in creating and deploying its AI model.
Why it matters: This case is significant as it seeks to establish liability for AI developers and the content generated by their models. The outcome could have implications for the accountability of tech companies and their role in preventing the creation and dissemination of non-consensual, degrading content. The case also highlights the need for greater responsibility and regulation in the development and deployment of AI technology.
3 million: the number of sexualized images generated by Grok in less than two weeks, according to researchers
Just as if you're an architect and build a building, you have liability for that architecture. Those that build and deploy AI models make design choices about how these models operate. This will be the case that looks at liability for decisions in those design choices. - Ravi Naik, legal director of AWO
The Guardian · 5 Jun, 05:00
Armed men abducted 39 pupils and seven teachers from schools in Oyo state, Nigeria, marking a spread of Chibok-style kidnappings to south-west Nigeria. The abductions have caused widespread outrage and concern about the country's insecurity challenges.
On May 15, armed men in military camouflage and face masks rode into the town of Yawota in Oyo state, Nigeria, on motorcycles and abducted 39 pupils and seven teachers, including a school principal, from schools in the area. The youngest children taken were two and three years old. Coordinated raids also took place in two other towns. A teacher, Joel Adegboye Adesiyan, was shot dead trying to protect his pupils. Two days after the kidnapping, a mathematics teacher and father of two, Michael Oyedokun, was beheaded in captivity.
Why it matters: The abductions are significant because they mark a spread of Chibok-style kidnappings, which were previously confined to northern Nigeria, to south-west Nigeria. This development has raised concerns about the country's insecurity challenges and the ability of the government to protect its citizens. The abductions have also forced citizens and the government to pay huge ransom demands, spawning a criminal sub-economy.
39 pupils and seven teachers abducted, 2 and 3 years old (youngest children taken), 276 schoolgirls abducted in Chibok in 2014, 2.57bn naira (£1.4m) estimated ransom paid to kidnappers in 2020, 1,000 forest guards to be recruited in Oyo state, 15 May (date of abduction), 2014 (year of Chibok abduction), 2020 (year of estimated ransom paid)
You don’t need force; all they have to do is negotiate with them and release us. We are in the cold, we are under the sun, we are under the rain. The children and the adults as well. Please, we are begging you, don’t let them waste our lives; they are getting impatient and frustrated. Please help us. We are begging you in the name of God. Please don’t just leave us, don’t forget us in the bush. - Alamu Folawe, head teacher of Ahoro-Esinele school
Ars Technica · 4 Jun, 21:25
The Supreme Court ruled that the FCC's process for issuing fines to AT&T and Verizon for selling users' location data without consent did not violate their right to a jury trial, upholding the fines. The carriers were fined a total of $104 million in 2024.
The Supreme Court ruled in favor of the Federal Communications Commission (FCC) in a case against AT&T and Verizon. The two telecommunications companies were fined $104 million in 2024 for selling users' real-time location data without consent. The carriers had challenged the fines, claiming that the FCC's process for issuing financial penalties deprived them of their Seventh Amendment right to a jury trial. The Supreme Court's decision, which was 8-1 with Justice Clarence Thomas dissenting, reversed a lower court ruling and upheld the fines.
Why it matters: This ruling is significant because it establishes that the FCC's process for issuing fines does not violate the right to a jury trial. It also upholds the fines imposed on AT&T and Verizon for selling users' location data without consent, which highlights the importance of protecting consumer privacy.
$104 million, 2024, 2018, 8-1
Ars Technica · 4 Jun, 20:02
Dashlane, a password manager provider, was targeted by a coordinated hacking campaign where attackers attempted to recover encrypted password vaults from its users. The attackers managed to download encrypted vaults from fewer than 20 personal user accounts before being shut down.
The attack began with a threat actor abusing Dashlane's mechanism for adding new devices to user accounts. They used a brute force attack to send a large volume of automated requests to Dashlane's API endpoints for device registration. This triggered an automatic lockout of targeted accounts, but before the attack was fully mitigated, the threat actor managed to brute force and generate valid tokens for fewer than 20 personal plan customers. This allowed them to register a new device on those accounts and download copies of users' encrypted vaults.
Why it matters: This incident highlights the vulnerability of password manager services to coordinated hacking campaigns. Although Dashlane's security systems were able to mitigate the attack and protect most user accounts, the incident shows that even established services can be targeted and potentially compromised. The encrypted vaults downloaded by the attackers still pose a risk if the encryption keys are compromised or cracked.
fewer than 20 personal user vaults, Sunday (start of the campaign), Thursday (update published by Dashlane)
The threat actor targeted the API endpoints for device registration and used a brute force attack to send a large volume of automated requests to those endpoints.
Ars Technica · 4 Jun, 20:44
The Estonian Language Institute (ELI) has released a 'Propaganda Resistance' benchmark ranking large language models (LLMs) on their ability to resist Russian propaganda. The benchmark evaluates LLMs' ability to avoid taking positions on topics that the Russian Federation uses in its strategic narratives.
The Estonian Language Institute (ELI) has developed a benchmark to evaluate the ability of large language models (LLMs) to resist Russian propaganda. The benchmark, called 'Propaganda Resistance', assesses LLMs' ability to avoid taking positions on topics that the Russian Federation uses in its strategic narratives. The researchers identified 14 broad categories of Russian influence operations and developed separate questions phrased to be neutral, biased with 'false assumptions' based on Russian propaganda, or to maliciously attempt to elicit explicit misinformation from the LLM. The questions were provided to the models in English, Estonian, and Russian.
Why it matters: This benchmark is significant because it helps to identify LLMs that are more resistant to spreading Russian propaganda, which is a concern for governments and citizens, particularly in countries like Estonia with a history of conflict with Russia. By evaluating LLMs' ability to resist propaganda, the benchmark can help to prevent the spread of misinformation and promote more accurate and reliable information.
NPR News · 4 Jun, 23:39
Summary not available.
NPR News · 4 Jun, 23:19
Summary not available.
NPR News · 4 Jun, 20:36
Summary not available.