AI risks News and Research

RSS

AI Doesn't Create Bias—It Reinforces It, Study Warns

AI bias isn't just a technical flaw—it's a reflection of societal power structures. Researchers argue that without transparency and accountability, AI will continue to reinforce discrimination rather than eliminate it.

27 Feb 2025

Governments Must Act Fast: AI Report Highlights Growing Dangers

The first International AI Safety Report, led by Yoshua Bengio, consolidates expert insights on AI risks and policy guidance for global decision-makers.

3 Feb 2025

Deepfake Democracy? AI Threatens 2025 Elections Worldwide

As AI reshapes political landscapes, researchers propose a four-point global framework to safeguard electoral integrity against AI-driven threats, ensuring fair and transparent democratic processes.

3 Feb 2025

AI Risks Escalate as Advanced Models Gain Power and Autonomy

AI safety experts warn that rapid advancements in general-purpose AI pose increasing risks, from cyber threats to systemic economic impacts, urging better safeguards and international oversight.

2 Feb 2025

How OpenAI Is Pioneering Safer AI Systems Through Hybrid Red Teaming

OpenAI advances AI safety with a dual red teaming approach, integrating human expertise and automated systems to uncover vulnerabilities and enhance model resilience. This innovative framework fosters public trust by addressing risks transparently and proactively.

3 Dec 2024

Anthropic Tests AI for Hidden Threats: Evaluating Sabotage Risks to Ensure Safe Deployment

Researchers from Anthropic explore AI's potential to secretly influence and manipulate decisions, highlighting the urgency for new evaluation strategies as AI capabilities advance.

23 Oct 2024

G7 Toolkit Empowers Governments to Deploy Ethical and Transparent AI

The G7 Toolkit for Artificial Intelligence in the Public Sector outlines strategies for ethical, secure, and effective AI deployment in governments, emphasizing human rights and transparency. It includes case studies and best practices to guide responsible AI adoption globally.

17 Oct 2024

MIT's AI Risk Repository Launches Database of 777 AI Risks

Researchers at MIT's AI Risk Repository created a comprehensive database categorizing 777 AI risks into two detailed taxonomies, providing a foundational tool for managing AI-related hazards.

4 Sep 2024

Researchers Propose Global AI Framework To Tackle Rapid Tech Challenges

Research paper examines the complexities of global AI governance, proposing a cautious approach to developing an international regulatory framework that balances innovation with ethical and societal needs.

4 Sep 2024

Regulating Advanced AI: The Role of KYC Schemes in Ensuring Safety

In a proposal, researchers emphasize the need for the US government to mandate Know-Your-Customer (KYC) schemes for AI compute providers, especially cloud service providers, to address emerging security and safety risks associated with advanced AI models.

25 Oct 2023

The AI Paradox: Empowering Innovation While Taming the Risks

Researchers explore the catastrophic risks of advanced AI development and provide strategies to mitigate them, including addressing malicious use, managing AI races, handling organizational risks, and controlling rogue AIs through safety measures and proactive measures.

3 Jul 2023