Plainly speaking, Artificial intelligence is no longer a fringe technology. It has become a core component of modern business, from customer service chatbots to complex data analysis.
We often treat the Large Language Models (LLMs) that are at the core of this technology as trusted black boxes. But like any software, they can be tampered with, manipulated, and turned against their creators. Understanding the ways an AI model can be compromised is the first step toward building a secure and resilient AI infrastructure. There are three primary ways that the authors want to cover, in which they will explain how an AI model can be compromised and made to act in unpredictable (or malicious!) ways:
Read more…
Source: Trend Micro
Sign up for the Cyber Security Review Newsletter
The latest cyber security news and insights delivered right to your inbox
Related:
- UK’s Sellafield nuclear waste processing plant fined £333K for infosec blunders
October 4, 2024
The outfit that runs Britain’s Sellafield nuclear waste processing and decommissioning site has been fined £332,500 ($440,000) by the nation’s Office for Nuclear Regulation (ONR) for its shoddy cybersecurity practices between 2019 and 2023. Sellafield, located in Cumbria, England, manages more radioactive waste than any other nuclear site in the world, and decommissioning work happening at ...
- Zimbabwe faces alarming rise in cyber attacks amid bank hacking
October 3, 2024
Zimbabwe has witnessed a significant surge in cyber attacks in recent months, with local entities, including banks, falling victim to hacking, the country’s Minister of Information Communication Technology, Tatenda Mavetera has revealed. Mavetera said the threat is also a local phenomenon, citing recent bank hacks in the country. Mavetera stated that cybercrime is not just a ...
- CISA flags major Ivanti security flaw – patch now
October 3, 2024
The US Cybersecurity and Infrastructure Security Agency (CISA) has added a known Ivanti bug to its Known Exploited Vulnerabilities (KEV) catalog, signalling that it’s being actively abused in the wild. The bug that was just added is an SQL Injection vulnerability, found this spring in the Core server of Ivanti Endpoint Manager (EPM) 2022 SU5 and ...
- Threat Awareness – Shifting Phishing Techniques & Tips for Staying Safe
October 3, 2024
Phishing remains a significant and ever-evolving cybersecurity threat, with recent data showing a 28% rise in attacks between Q1 and Q2 of 2024. This trend highlights how persistent and evolving phishing tactics continue to be, impacting a staggering 94% of cybersecurity decision-makers in 2023. Attackers are increasingly using compromised internal accounts, shifting the platforms they ...
- Russian Authorities Arrest 96 in Major Money Laundering Operation
October 3, 2024
In a coordinated effort against cybercrime-related money laundering, Russian authorities have made nearly 100 arrests in connection with an extensive criminal operation involving cryptocurrency exchanges and illegal financial activities. The arrests were part of a nationwide crackdown tied to the UAPS payment system and the Cryptex cryptocurrency exchanges, both of which have been linked to cybercriminals ...
- News agency AFP notifies French authorities of potential data breach
October 2, 2024
Agence France-Presse (AFP), one of the world’s largest news organizations, has notified French regulators of a potential data breach following a cyberattack last week. The AFP, which has an editorial presence in 260 cities across 150 countries, said in a brief statement on Saturday that it detected an “attack on its systems” that affected part of ...