DeepMind launches institute to explore AGI’s technical and societal implications
Shane Legg says the DeepMind Institute aims to spur interdisciplinary research, collaboration and debate on how to build and govern artificial general intelligence safely.
TLDR
In his September 16, 2026 announcement, Shane Legg says artificial general intelligence is approaching. He acknowledges that current AI still sometimes fails at basic tasks and lacks the consistency and creativity needed for full AGI, but expects those gaps to close soon. The DeepMind Institute aims to explore the technical and societal questions this raises. Legg says AGI could accelerate advances in medicine, clean energy and prosperity, while highlighting cybersecurity and biological risks and the potential for loss of control over future self-improving systems.
Combined views
26.7K
1 Source, first seen 8d ago
DeepMind launches institute to explore AGI’s technical and societal implications
Shane Legg says the DeepMind Institute aims to spur interdisciplinary research, collaboration and debate on how to build and govern artificial general intelligence safely.
TLDR
In his September 16, 2026 announcement, Shane Legg says artificial general intelligence is approaching. He acknowledges that current AI still sometimes fails at basic tasks and lacks the consistency and creativity needed for full AGI, but expects those gaps to close soon. The DeepMind Institute aims to explore the technical and societal questions this raises. Legg says AGI could accelerate advances in medicine, clean energy and prosperity, while highlighting cybersecurity and biological risks and the potential for loss of control over future self-improving systems.