Tag: technews

Google employees label AI chatbot ‘Bard’ as worse than useless and a pathological Google employees have criticized the company’s chatbot Bard, citing it as a pathological liar and worse than useless. Despite the risk evaluations submitted by an internal safety team, Google opened up early access to the bot in March. This illustrates how Google…

Goldoson Android Malware Infects Over 100 Million Android Devices Malware strain Goldoson has been found in over 60 apps on the Google Play Store, with a combined 100 million downloads. An additional 8 million downloads were tracked from a third-party app store in South Korea. The malicious library is capable of gathering information about installed…

YouTube Videos Distributing Malware via Highly Evasive Loader Cybersecurity researchers have uncovered a new malware loader, in2al5d p3in4er, which is used to deliver the Aurora information stealer malware. The malicious software is distributed through YouTube videos and fake cracked software download websites. The loader is designed to evade detection by querying the vendor ID of…

Leveraging Mobile Phone Data with Machine Learning to Target and Fight Poverty Researchers from UC Berkeley and the World Bank have developed a machine learning model using non-traditional administrative data (such as call detail records) to accurately target ultra-poor households for government anti-poverty programs in low to medium-income countries. The model was found to have…

Some Glimpse of AGI in ChatGPT. Others Call It a Mirage Microsoft researchers have recently made claims that their new AI system, GPT-4, may be exhibiting sparks of artificial general intelligence. GPT-4 is a powerful text generation algorithm that is capable of creating scientific diagrams, writing poetry and prose, and solving coding problems. Although it…

Meta Launches Horizon Worlds social VR Meta is launching their Horizon Worlds social VR platform to teens in the US and Canada in the coming weeks. To ensure a safer experience, Meta has implemented features such as private profiles, voice mode to turn voices of people you don’t know into “quiet, friendly sounds”, and expanded…

Reddit Changes API Today, Reddit announced changes to its API that will limit the use of data from its content pipeline to train artificial intelligence tools. This will require commercial users to enter into a separate agreement with Reddit in order to use the data. It is unclear how much this agreement will cost. Reddit’s…

Apple macOS Devices Are Now the Target of LockBit Ransomware The LockBit ransomware crew has developed a macOS-based payload that has been available since November 11, 2022, and has managed to evade detection by anti-malware engines until now. It is the second most used ransomware in March 2023, and its new macOS version relies on…

Apple releases security updates for iOS, iPadOS, macOS, and Safari to address zero Apple released security updates on April 10, 2023 to address two zero-day vulnerabilities that were being actively exploited. The first was a use after free issue in WebKit that could lead to arbitrary code execution and the second was an out-of-bounds write…

The Verge of something The Verge of a new era The Verge of Elon Musk has proposed creating a new AI system called TruthGPT in order to avoid the potential destruction of humanity. The AI would be designed to understand the universe and its ethical implications, and Musk has compared its potential benevolence to the…