AI Safety: Building Safer and More Responsible AI Systems
AI systems are becoming increasingly capable and widely deployed, making AI safety an important question for developers, researchers, and society.
Join the PyTorch East Africa Community for an online session with Jabess Omane, an AI Research Engineer from Accra, Ghana, whose work focuses on AI safety, interpretability, alignment, and responsible AI.
In this session, we will explore the foundations of AI safety and some of the key challenges facing modern AI systems, including:
Bias and fairness
Misuse and security risks
Model unreliability and hallucinations
Opacity and interpretability
Alignment with human values
Emerging questions around open-weight and closed AI models
How developers and open-source communities can contribute to safer AI
Jabess leads the technical work of the Africa AI Sovereignty Network (AISN) in interpretability and contributes to the Africa AI Sovereignty Index (AISI) methodology. He is also a Thinking About Thinking Ambassador and a Ghana School of Internet Governance Fellow 2026.
The session aims to make AI safety accessible, practical, and relevant to practitioners in East Africa and beyond, while creating space for discussion around how we can build and deploy AI systems that are safer, more responsible, and trustworthy.
💻 Online
🎙️ Speaker: Jabess Omane
📅 Date: August 29
🕐 Time: 9:00 PM (EAT)
🌍 Open to the AI/ML community
Join the community group: Join WhatsApp Group
Follow us on LinkedIn: pytorch-east-africa