We do not control neural networks well enough to guarantee an AI will not harm humans, so we should be careful about what capabilities we give them.

Added to korrents 29 Aug 2026 · How quotes work · Something wrong? Tell us

Shared subject first, then similar wording. They may agree or disagree.

• A neural network's latent space is closer to an uncopyrightable syntax than to copyrightable expression. KKKevin Kelly7 Sept 2026neural networks
• A verifiable task can be optimised by reinforcement learning until a neural network performs it extremely well. AKAndrej Karpathy17 Nov 2025neural networks
• Access to large-scale generalist AI systems that could be weaponized should be limited, meaning their code and neural network parameters should not be released open-source. YBYoshua Bengio24 Jun 2023neural networks
• AI is unlikely to be a hyperprofits engine for research labs whose attached companies make money serving neural-network calculations from datacenters. BDBrad DeLong21 Sept 2026neural networks
• An agent can understand a maximally simplified explanation and still be unable to come up with it — that gap is what is left of the expert's job. AKAndrej Karpathy20 Mar 2026neural networks

See this claim on the map