2024-05-19
Subscribe to get the latest
Digital Transformation for May 19, 2024 includes news in artificial intelligence, cybersecurity, and edge computing. This week cybersecurity threats at the edge and new jail breaking LLM techniques revealed.
Keywords
#ai #compute #edge #cybersecurity #datamanagement #advancedcomms #edgeplatform #intelai #edgeai #aiapplications #iotdevices #aiot #aiotedge #edgecybersec #iotsecurity #usfcc #uscybertrust #ibmqradar #entedge #bankingtrojans #aisafety #ailegislation #aielection #deepfake #cyberthreat #largetextmodel #llm #chatgpt #llama3 #digitaltrans #embracindigital #robbertvaughn #intelit
Edge Computing
Intel has launched a new Edge Platform to streamline the development and management of edge AI applications. The modular, open software platform supports enterprises in deploying and securing AI solutions efficiently, enhancing the total cost of ownership. Key features include simplified infrastructure management, advanced AI capabilities, and seamless integration with existing systems. The platform, backed by industry leaders like AWS, Lenovo, and SAP, addresses complexities in edge computing, ensuring optimal performance and security. Read More
The World Economic Forum highlights the critical role of the Internet of Things (IoT) in modern supply chains, emphasizing how IoT devices enhance efficiency and resilience. Despite the rise of cloud computing, edge computing remains vital for real-time operations and security. Fusing AI with IoT (AIoT) drives innovation, promising a more connected and efficient future. This technology is crucial for managing vast amounts of data and improving operational responses in various industries Read More
Edge computing offers significant benefits but comes with five major cybersecurity risks: IoT-specific vulnerabilities, an overwhelming amount of logs, data compromises, a broad attack surface, and budget limitations. Strategies to mitigate these risks include using authentication controls, automating log monitoring with AI, authenticating devices and users, encrypting network traffic, and deploying AI-driven intrusion detection systems. Read More
Cybersecurity
The US Federal Communications Commission (FCC) has approved a voluntary cybersecurity labeling program for wireless consumer IoT products. The initiative aims to help consumers identify secure devices through a “US Cyber Trust Mark.” This label, accompanied by a QR code with security details, incentivizes manufacturers to enhance cybersecurity standards. The program will involve public-private collaboration, with the FCC overseeing and third-party administrators managing the labeling process. The move addresses growing cybersecurity concerns as the number of IoT devices rises. Read More
IBM’s sudden exit from its cybersecurity software division has left many CISOs scrambling for alternatives. The unexpected move affects IBM’s QRadar SaaS users, a leading security information and event management (SIEM) solution. Organizations must quickly transition to other providers, which could potentially lead to security gaps during the migration period. This highlights the critical need for contingency planning in cybersecurity strategies to ensure data protection during such transitions. Read More
Cybercriminals are exploiting GitHub and FileZilla to spread malware, including stealer malware and banking trojans like Atomic, Vidar, and Lumma. They use fake profiles and repositories to distribute malicious software, targeting Android, macOS, and Windows platforms. The attack leverages legitimate services and involves sophisticated techniques such as SEO poisoning and malvertising. This highlights the growing misuse of trusted platforms for cyberattacks and underscores the need for heightened cybersecurity measures. Read More
Artificial Intelligence
The world addresses AI risks through various initiatives, including the AI Safety Summit in Korea, the EU’s AI Act, the AI Pact, and international efforts by the OECD, G7, and the United Nations. These measures aim to regulate AI’s impact on society, promote responsible innovation, and ensure human rights. Key developments include risk-based AI legislation, global safety commitments, and collaborative frameworks to manage AI’s ethical and operational challenges. Read More
The United States Department of Homeland Security has sounded the alarm on AI election threats, highlighting the potential risks of AI-generated disinformation, cyberattacks, and deepfake technology. This urgent call to action emphasizes the increasing vulnerability of elections and the need for bolstered cybersecurity measures. Launching public awareness campaigns is crucial to protect democratic processes, underscoring the immediate need to address AI’s role in election security. Read More
Researchers have raised concerns about the safety and robustness of Large Language Models (LLMs) like ChatGPT, Llama3, and Gemini. A study by researchers at AWS AI Labs has revealed vulnerabilities to adversarial attacks and jailbreaking, with successful attacks reaching 90% in the current models. They have proposed countermeasures that promise to reduce the success of such attacks. The models achieve top performance in spoken question-answering tasks, with safety and helpfulness metrics exceeding 80%. Read More
Embracing Digital Transformation
Darren is continuing his discussion with Robert Vaughn about how companies, both big and small, are using Intel’s IT department to learn best practices and elevate the industry as a whole. Also, don’t forget to share the Embracing Digital Transformation podcast with your colleagues. The weekly listenership is continuing to grow, with over 30,000 weekly listeners. Read More