The Trump administration has rebranded the AI Safety Institute to the Center for AI Standards and Innovation, signaling a shift towards rapid technology development. Commerce Secretary Howard Lutnick emphasized that the center will continue to evaluate AI capabilities and vulnerabilities while promoting U.S. innovation.
Two Republican senators expressed their support for continued federal investment in artificial intelligence research at the AI + Expo in Washington, D.C. They emphasized the importance of public funding for innovation and collaboration to maintain U.S. leadership in the AI sector.
Despite numerous AI governance events, a comprehensive framework for AI safety has yet to be established, highlighting the need for focused dialogue among stakeholders. A dual-track approach that combines broad discussions with specialized dialogue groups could foster consensus and address context-specific risks effectively.
The article emphasizes the urgent need for the U.S. to adopt a “black box” system for AI, similar to aviation, to learn from failures and enhance safety and governance in AI technologies. It advocates for improved AI literacy among the population to ensure that Americans can navigate the complexities of an AI-driven economy effectively.
An internal briefing note for Canada’s election watchdog warns that the use of artificial intelligence poses a “high” risk for the ongoing election campaign. It highlights concerns about potential violations of the Elections Act, particularly related to disinformation and the use of AI tools to mislead voters.
Artificial Intelligence Safety Levels (ASLs) categorize AI safety protocols into distinct stages, ranging from ASL-1 with minimal risk to ASL-4 where models may exhibit autonomous behaviors. Currently, we are at ASL-2, and there is an urgent need for regulations to address the potential risks associated with advancing AI capabilities.
The paper discusses the challenges of moral accountability and safety assurance in the use of artificial intelligence-based clinical tools in healthcare. It emphasizes the need to update our understanding of accountability due to the opaque decision-making processes of these systems and suggests involving AI developers in the assessment of patient harm.
The European Union has passed its first major Act to regulate artificial intelligence, aiming to ensure safety and protect fundamental rights. The legislation introduces a risk-based approach that categorizes AI systems and applies stricter regulations to those deemed high or unacceptable risk.