A new loop denial-of-service attack vector has recently been discovered by CISPA Helmholtz Center for Information Security researchers that uses vulnerabilities affecting Broadcom, Microsoft and Honeywell products, as well as approximately 300,000 impacted internet hosts. This method can reportedly let a hacker render a targeted service unusable or cripple a network's backbone and cause a network outage, and it can also be used for DoS or DDoS attack amplification.
Google Health is expanding its reach into health care AI, with a version of its Gemini large language model adjusted for the sector, while its Articulate Medical Intelligence Explorer tool is undergoing testing at one health care facility to guide diagnoses and facilitate communication with patients. Google is also working with Fitbit to develop an LLM for Fitbit devices and is rolling out an AI-based feature that allows users to visually search for health conditions.
Two AI image generators were found to produce racially and sexually biased results and amplify stereotypes, and using the tools to populate websites or for medical pamphlets will worsen biases, researchers say. Part of the problem lies with how training images are annotated, say AI research scientists Kathleen Fraser and Pratyusha Ria Kalluri. Simply writing better prompts or explicit counter-prompts hasn't been shown to solve the problem, and automated filtering based on keywords has had unintended effects.
An "AI-first" approach in the workplace can backfire by making employees feel like they're a lower priority than technology and putting an emphasis on algorithms instead of human behavior, writes Oguz A. Acar from the King's Business School at King's College London. "Instead of embracing an AI-first strategy, I recommend organizations keep the 3Ps at the forefront of their AI transformation: problem-centric, people-first, and principle-driven," Acar writes.
Eager to upgrade its mobile AI offerings, Apple has had discussions with startups including OpenAI and Cohere, as well as Google and other providers, sources said. Apple wants the next version of its iPhone operating system, iOS 18, to have more generative AI features, they said.
A federal judge has sided with Google, dismissing a 2020 lawsuit from two residents of Illinois accusing the search giant of gathering 100 million photos of faces in violation of the state's Biometric Information Privacy Act. The reason given for the dismissal was that the plaintiffs failed to prove any of Google's activities took place in Illinois.
The Biden Administration announced a landmark funding deal with Intel worth up to $19.5 billion, including $8.5 billion in direct funding and additional loan support. Part of a broader push through the CHIPS and Science Act, the funds are slated to elevate US semiconductor production capabilities with four new manufacturing sites.
Good cybersecurity people don't need to strive for perfection. There will inevitably be new and challenging incidents to manage. Although having technical aptitude and a good understanding of threats and regulations are great, that can all be learned -- but it isn't enough. Here are some essential traits to look for when hiring good cybersecurity staff.
These days organizations are faced with a higher probability of facing a cyberattack, and attacks can often cause a great deal of damage to an organization's reputation and operations. This is why organizations must write and implement acceptable use policies -- regardless of size. Learn what a corporate acceptable use policy is, the key areas that should be included and some pro tips to enhance your security awareness training.