AI Safety

Artificial intelligence news, analysis and reporting on AI Safety from AI Balkans.

Latest on AI Safety

Open-Weight AI Models Are Reaching Advanced Cyber Capabilities — Anthropic and NIST say Z.ai’s GLM-5.3 brings advanced cyber capabilities to an openly downloadable AI model.
AI Summaries Can Change What People Remember — AI-generated summaries containing incorrect information can alter what people later remember about an event, according to new research from Georgetown…
AI Leaders Question the Race to Go Faster — As AI companies debate slowing frontier development, businesses and governments are still racing to adopt more AI. What does that gap mean?
Google Earth’s AI Experiment Lasted One Day — Google removed an AI image tool from Google Earth after users created fictional scenes over real locations, raising misinformation concerns.
UN Opens First Global AI Governance Dialogue in Geneva — The UN has opened its first Global Dialogue on AI Governance, calling for safer, more inclusive AI rules as smaller regions seek a stronger voice.

About this topic

5 stories since July 2026 on AI Safety. Often covered together with AI, Anthropic and AI Adoption. Most coverage sits in Global AI, Policy & Regulation and Analysis. Latest: Open-Weight AI Models Are Reaching Advanced Cyber Capabilities.

Timeline

6 July 2026 UN Opens First Global AI Governance Dialogue in Geneva
1 August 2026 Google Earth’s AI Experiment Lasted One Day
14 September 2026 AI Leaders Question the Race to Go Faster
26 September 2026 AI Summaries Can Change What People Remember
1 October 2026 Open-Weight AI Models Are Reaching Advanced Cyber Capabilities

AI Safety

Latest on AI Safety

About this topic

Related topics

Timeline

Latest articles