AI & Data Engineering News
Anthropic alleges Chinese AI firms routed user queries to Claude
Anthropic claims in a new report that Chinese AI developers Moonshot AI and DeepSeek routed their users' queries to its Claude models while presenting the answers as output from their own systems....
AI-powered attack exploited PaperCut flaws to hack 395 organizations
A threat actor, likely Russian-speaking, used hundreds of AI agents to develop and launch a global exploitation campaign targeting vulnerable PaperCut NG/MF servers. [...]
IDScan confirms breach tied to 153 million stolen driver’s licenses
Identity verification company IDScan has confirmed that hackers accessed customer data stored in its cloud platform, days after reports linked the company to a massive database containing more than...
More questions about whether researchers can trust OpenAI with unpublished math
https://mathstodon.xyz/@andreasthom/117240536885387540 https://mathstodon.xyz/@andreasthom/117240537520615623 https://x.com/ValerioCapraro/status/2097791836269977996 ...
OpenAI's Astra model marks first to hit 'critical' cybersecurity threshold
OpenAI announced Tuesday that its upcoming AI model, Astra, is the first to cross the company's "Critical" cybersecurity capability threshold. According to the firm, Astra can identify and exploit...
Micron leads our AI memory stock picks as DRAM, NAND, and HDD markets diverge
Micron, Sandisk, Seagate, and Western Digital have become four of the hottest AI investment plays this year, all ranking among the S&P 500's top 10 performers. Sandisk leads with a roughly 575%...
U.S. Targets Cloud Loophole in Nvidia Chip Export Ban
U.S. export controls on Nvidia's most powerful AI chips are meant to keep China from accessing cutting-edge compute, but several Chinese firms have apparently sidestepped the restrictions by tapping...
ChatGPT teen accounts get new safety controls, including a less human voice
OpenAI is introducing a set of safety features for under-18s using ChatGPT, starting today. Teenagers with new accounts will be able to turn off the human-like voice response, reducing how much the...
Nvidia leans on its balance sheet to keep the AI boom rolling
Nvidia's early lead in artificial intelligence made it the world's most valuable company, but rivals like AMD and Google have narrowed the technology gap. Now the chipmaker is leaning on a different...
Nvidia Backs $105B Ohio AI Data Center for OpenAI
Nvidia will provide up to $105 billion in financing for a new AI data center for OpenAI in Ohio, according to a securities filing made public on Monday. The credit supports an initial 4.25 gigawatts...
Alibaba counters Meta with laptop-friendly AI model and open weights
Alibaba launched a new AI model on Monday designed to run on consumer hardware like laptops, directly responding to Meta's recent push into open-source AI. The Chinese tech giant also released the...
Meta and Nvidia stake claim in open-weight AI race with new model releases
Two American tech giants are pushing back against calls for restrictions on open-weight AI by releasing their own free-to-download models this week, aiming to compete with popular offerings from...
Nvidia teams with Wall Street giants on $500bn AI infrastructure financing push
Nvidia has signed agreements with several major Wall Street firms to raise over $500bn for the datacentres, chip plants and power infrastructure underpinning the AI boom. The company's CEO, Jensen...
OpenAI broadens Daybreak cyber program with new tiers and a dedicated model
OpenAI has expanded its Daybreak cybersecurity initiative, introducing two access levels and a new AI model tailored for defensive and offensive security work. The program, first launched in May, is...
Meta open-sources flagship AI model, unveils laptop-ready family in push against rivals
Meta announced it will open source its latest and most powerful AI model, making its weights publicly downloadable, while also introducing a new family of lightweight models designed to run on...
OpenAI restricts new model access over potential cyberattack capabilities as AI safety scrutiny grows
OpenAI has paused certain internal work involving an unreleased AI model, citing concerns that it might be capable of autonomous cyberattacks. The move comes amid a string of security incidents at...
Anthropic's Mythos model faked identities to push malicious code in AI safety test
Anthropic's Mythos model created fake online identities to pressure a real person into approving malicious code updates to an open source project, according to a new report from the U.K.-based AI...