Security
Hackers claim oil giant Petrobras, alleging oil-rich maps theft
A Russia-linked ransomware gang claims it has snatched 90GB of sensitive seismic and exploration data from Petrobras, pressuring the Brazilian petroleum giant to negotiate.
Read more about Hackers claim oil giant Petrobras, alleging oil-rich maps theft
AIPAC discloses data breach, customer information stolen
The American Israel Public Affairs Committee (AIPAC) has announced that it has been struck by a cyberattack, in which personally identifiable information (PII) was exfiltrated.
Read more about AIPAC discloses data breach, customer information stolen
Eurofiber France discloses data breach
The French division of Eurofiber has suffered a data breach in which some customer information was exfiltrated. In the meantime, all systems have been secured.
Read more about Eurofiber France discloses data breach
New TOAD phishing campaign targets Microsoft Entra guest invitees with fake invoices
Cybercriminals are targeting recipients of Microsoft Entra B2B guest invites, taking advantage of Entra’s cloud-based infrastructure to bypass email filters in a new Telephone-Oriented Attack Delivery (TOAD) phishing campaign.
Read more about New TOAD phishing campaign targets Microsoft Entra guest invitees with fake invoices
430k Cabify drivers’ records stolen, hackers claim
The allegedly stolen details include a wide range of personal information, ranging from home addresses to Facebook account IDs. Researchers believe exposed drivers could be targeted in social engineering attacks.
Read more about 430k Cabify drivers’ records stolen, hackers claim
US lawmakers suggest banning VPNs to protect children
Legislators in Wisconsin and Michigan are considering implementing laws that prohibit the use of Virtual Private Networks (VPNs) to visit websites that host content that’s harmful to minors.
Read more about US lawmakers suggest banning VPNs to protect children
Your questions, answered by Cybernews: Which certificate can land you a job in cybersecurity?
Certifications, hacking experience, or being a unicorn? What’s really needed to kick off your career in cybersecurity? Every week, our team selects one pressing and common reader issue and deconstructs it to help you stay safe online.
Read more about Your questions, answered by Cybernews: Which certificate can land you a job in cybersecurity?
The Washington Post reveals thousands impacted via Oracle-based hack
Attackers were able to access data owned by the Jeff Bezos-owned newspaper for over a month. Meanwhile, the hacks’ perpetrators accessed personal and sensitive information, which includes bank account numbers.
Read more about The Washington Post reveals thousands impacted via Oracle-based hack
With blazing-fast WiFi 7 speeds come extra security risks: what to know before upgrading?
One of the coolest features of WiFi 7 is its ability to combine and simultaneously utilize three frequency bands – 6 GHz, 5 GHz, and 2.4 GHz – achieving blazing speeds. However, if you consider a major network upgrade, you will now be managing three separate radio configurations. Misconfigure one of the bands, and the hackers might get in.
Read more about With blazing-fast WiFi 7 speeds come extra security risks: what to know before upgrading?
Cornerstone staffing ransomware attack leaks 120,000 resumes, claims Qilin gang
Cornerstone staffing, a nationwide US employment agency and tech recruiter, is allegedly breached by the Qilin ransomware group, which claims to have pilfered the “personal resumes” of over 120,000 job seekers, along with over 1 million company files.
Read more about Cornerstone staffing ransomware attack leaks 120,000 resumes, claims Qilin gang
DoorDash suffers another data breach: users’ contact information affected
Users of DoorDash, a major American food delivery company, are receiving emails informing them of a potential cybersecurity breach that may affect their data.
Read more about DoorDash suffers another data breach: users’ contact information affected
Massive blow to cybercrime: three malware families disrupted, 1025 servers taken down
Europol has achieved another major breakthrough in disrupting cybercriminal networks. Three notorious malware families – infostealer Rhadamanthys, trojan VenomRAT, and the botnet Elysium – have been crippled.
Read more about Massive blow to cybercrime: three malware families disrupted, 1025 servers taken down
Hackers claim to sell internal Samsung subsidiary files
A hacker is attempting to sell what they claim is internal Samsung data stolen through a compromised contractor.
Read more about Hackers claim to sell internal Samsung subsidiary files
Akira updates its arsenal, extorts $250 million in ransomware proceeds
Akira, the second most active ransomware clan, has expanded its capabilities. It has introduced a new ransomware variant and has been attacking small and medium-sized businesses, claiming over 620 victims this year.
Read more about Akira updates its arsenal, extorts $250 million in ransomware proceeds
Chrome Web Store still features a crypto-stealing extension despite takedown request
App marketplaces and browser extension stores are increasingly known for allowing criminals to publish malicious software. The most recent example comes from the Chrome Web Store.
Read more about Chrome Web Store still features a crypto-stealing extension despite takedown request
New "Scam Center Strike Force" to combat crypto investment fraud centers run out of Southeast Asia
The US Justice Department (DoJ) has formed a new “Scam Center Strike Force" to go after online criminal groups operating in Southeast Asia that target Americans with promises of "get-rich-quick" crypto schemes – known as pig butchering – already raiding two centers and confiscating over $400 million in digital currency.
Read more about New "Scam Center Strike Force" to combat crypto investment fraud centers run out of Southeast Asia
We tested ChatGPT, Gemini, and Claude with adversarial prompts: here are our findings and risks
A Cybernews comparative study probes ChatGPT, Google Gemini, and Anthropic Claude using adversarial prompts. Results reveal varied safety gaps, common bypass techniques, and action points for developers.
Read more about We tested ChatGPT, Gemini, and Claude with adversarial prompts: here are our findings and risks
Which AI model would be Pablo Escobar’s best friend?
Not all large language models (LLMs) were smart enough to avoid revealing the intricacies of manufacturing a Schedule II controlled substance, Cybernews researchers have discovered.
Read more about Which AI model would be Pablo Escobar’s best friend?
Google’s Gemini 2.5 Pro has a serious safety problem
While Google’s Gemini 2.5 Flash is among the safest models when it comes to resisting malicious prompts, its Gemini 2.5 Pro can be easily tricked into generating detailed animal abuse methods, advice on stalking, and other questionable content.
Read more about Google’s Gemini 2.5 Pro has a serious safety problem
Google backs off harsh sideloading restrictions after community backlash
Google is softening its approach towards the new Android developer verification requirements after widespread pushback. Most Android users will still be limited to downloading apps from only verified developers. However, “experienced users” will get a workaround.
Read more about Google backs off harsh sideloading restrictions after community backlash
