Author: user
Ravi LakshmananJanuary 22, 2026Network security/vulnerabilities Cybersecurity firm Arctic Wolf warned of a “new cluster of automated malicious activity” involving unauthorized firewall configuration changes on Fortinet FortiGate devices. The group said the campaign began on January 15, 2026, adding that it bears similarities to a December 2025 campaign in which malicious SSO logins on FortiGate appliances were recorded against administrator accounts from various hosting providers, exploiting CVE-2025-59718 and CVE-2025-59719. Both vulnerabilities could allow unauthenticated bypass of SSO login authentication via a crafted SAML message if the FortiCloud Single Sign-On (SSO) feature is enabled on an affected device. This shortcoming affects FortiOS,…
Rabi LakshmananJanuary 22, 2026Vulnerability/Zero-day Cisco has released a new patch to address what it describes as a “critical” security vulnerability affecting multiple unified communications (CM) products and Webex Calling dedicated instances. This vulnerability is actively being exploited in the wild as a zero-day attack. Vulnerability CVE-2026-20045 (CVSS score: 8.2) could allow an unauthenticated, remote attacker to execute arbitrary commands on the underlying operating system of an affected device. “This vulnerability is due to improper validation of user-supplied input in an HTTP request,” Cisco said in an advisory. “An attacker could exploit this vulnerability by sending a series of crafted HTTP…
In the world of AI infrastructure, a pattern is emerging of popular open source tools turning into venture-backed startups worth hundreds of millions of dollars. The latest example is RadixArk, the for-profit company behind SGLang, a popular tool that allows AI models to run faster and cheaper. RadixArk was recently valued at about $400 million in a funding round led by Accel, according to two people familiar with the matter, a remarkable amount for a startup that was only announced last August. TechCrunch was unable to confirm the size of the funding. The news comes as part of the team…
Bluesky’s “Starter Pack” has proven to be a popular way to help people connect with others on social networks, with a curated list of people to follow. In fact, it’s so popular that X is now copying this feature. On Wednesday, Nikita Via, head of product at Company X, announced in a post that the Elon Musk-owned app will soon be introducing its own version of these lists, which it calls a “starter pack.” (How ingenious!) The idea behind the new feature is to allow users to find accounts that match their interests across a variety of categories, including news,…
You might not need an AI smart ring to manage your to-dos. Doist, the company behind Todoist, the popular to-do list app that works on multiple devices, is releasing Todoist Ramble, a new AI-powered “voice to task” feature. This feature allows you to describe tasks using natural language. As the name suggests, Ramble can turn your meandering, unstructured speech into organized tasks. The app also captures the other details you mentioned, such as project deadlines, priorities, duration, and assignees. Image credit: Todoist People often think about things to do while on the go, but it can be difficult to pull…
Apple’s long-awaited Siri revamp could turn the smart assistant into a chatbot similar to ChatGPT, Bloomberg’s Mark Garman reports. His sources say the Siri chatbot integrated into iOS 27 could be the focus of Apple’s WWDC presentation in June. The Siri chatbot, internally codenamed “Campos,” works with both voice and text input. Apple senior vice president Craig Federighi has previously said he doesn’t want Siri to be a chatbot, but rather integrates Apple’s AI options so that they’re always within reach when you need them. But with pressure mounting from the success of other companies’ AI chatbots, the company appears…
On Wednesday, Antropic released a revised version of the Claude Constitution. This is a living document that provides a “holistic” explanation of “the context in which Claude operates and the kind of existence we want him to have.” The document was released on the sidelines of Anthropic CEO Dario Amodei’s appearance at the World Economic Forum in Davos. Anthropic has long differentiated itself from its competitors through a system it calls “Constitutional AI.” It’s a system in which the company’s chatbot, Claude, is trained using specific ethical principles rather than human feedback. Anthropic first published those principles, the Claude Constitution,…
Online advice to “touch grass” to calm your emotional state may be backed up by science, at least in lab mice.A recent study found that mice that live outdoors are less anxious than mice that spend their days in secure shoebox-sized cages. And that could ultimately highlight fundamental flaws in laboratory research, including research used to test the safety and effectiveness of drugs in humans.Drugs that seem to work in lab mice don’t necessarily work in human patients, and some scientists believe that drugs can fail because of how strange and isolating life in the lab is for rodents. you…
European consumers are firing back at the United States after President Trump threatened to seize control of the Danish territory of Greenland. As a result, two mobile apps that offer ways to see if a product is American-made and suggest local alternatives have rocketed to the top of the Danish App Store in recent days. The Danish App Store serves users in both Greenland and Denmark, and the same is true for Google Play. The increase in downloads comes as Danish consumers have organized a grassroots boycott of American products, including canceling vacations in the United States and dropping subscriptions…
AI detection startup GPTZero scanned all 4,841 papers accepted to the prestigious Neural Information Processing Systems Conference (NeurIPS) held in San Diego last month. The company discovered 100 hallucinatory citations across 51 papers and confirmed them to be fake, the company told TechCrunch. Having a paper accepted by NeurIPS is a resume-worthy accomplishment in the world of AI. Given that they are leading experts in AI research, one might assume that they would use LLM for the devastatingly boring task of writing citations. Therefore, this finding requires a lot of caution. The 100 identified hallucination citations across 51 papers are…