Article AI

AI News Today: Security Threats, Governance Gaps, and Privacy Concerns Dominate

Today's AI news highlights mounting security concerns, unanswered legal questions, and ongoing privacy debates surrounding artificial intelligence systems.

AI-Designed Viruses Test Biosecurity Limits

Researchers are exploring how artificial intelligence can be used to design viruses, raising urgent questions about whether existing biosecurity frameworks can keep pace with rapidly advancing AI capabilities in biotechnology. Read more: AI-designed viruses are a test of whether biosecurity can keep pace - The Conversation

Taiwan Hit by AI-Powered Cyber Attacks

Financial sector and telecommunications organizations in Taiwan were hit by a sophisticated cyber intrusion attributed to China-linked actors, in what researchers are calling an unprecedented use of autonomous AI capabilities in the attack chain. Read more: China-linked hackers hit Taiwan in unprecedented 'autonomous' AI cyber attack - Financial Times

Taiwan Discloses AI-Driven Hacking Campaign

Taiwanese authorities have disclosed that the island was targeted by AI-driven cyberattacks last month, marking an escalation in state-sponsored hacking operations that leverage artificial intelligence agents. Read more: Taiwan says it was targeted last month in AI-driven hacking campaign - Reuters

Chinese Censorship Patterns Surface in American AI

Studies indicate that Chinese government censorship requirements are surfacing in responses from American AI models, raising concerns about information integrity and the global reliability of AI systems. Read more: Chinese Censorship Is Leaking Into Answers From American AI - WSJ

Twitch Under Fire for AI Training on User Content

Twitch is under fire after users discovered that Amazon, its parent company, has been using livestream content from the platform to train AI models, sparking privacy and consent concerns. Read more: Twitch faces backlash over Amazon training AI on its users' livestreams - BBC

Legal Accountability Gap for Autonomous AI Agents

As AI agents gain autonomy in executing complex tasks, legal experts warn that existing frameworks have no clear mechanism to hold anyone accountable when these systems cause harm to people or property. Read more: AI agents aren't legally responsible for any harm that they cause, experts say. So who is? - The Guardian

OpenAI Introduces 'Dots,' Always-On AI Agents with Dedicated Cloud Resources

OpenAI has unveiled 'Dots,' a new feature introducing always-on AI agents directly within ChatGPT. These agents are designed to operate continuously, providing persistent AI assistance and interaction. A notable aspect of 'Dots' is that each agent is provisioned with its own dedicated cloud computing resources. This

White House Issues Executive Order on 'Super Intelligence' Amid New Industry Accord

The White House has unveiled a new executive order focused specifically on what officials are calling "super intelligence"—highly advanced AI systems that may approach or exceed human cognitive capabilities in key domains. The order arrives as leading AI companies continue to push the boundaries of model capabilities,

White House Publishes Short AI Accord Involving Tech Leaders

The White House has made public a compact agreement reached with several prominent AI executives, with the documented text spanning just 308 words. The accord represents a voluntary framework addressing key concerns around artificial intelligence development, though details of the specific commitments remain limited to the official document. Such short-form