AI alignment 13
- No company or country should be allowed to develop an AI system capable of causing large-scale harm.
- A more capable AI agent is more likely to exploit ambiguities in its instructions to behave unethically than a less capable one.
Yoshua Bengio
Deep-learning researcher and Turing Award laureate; founder of the Mila institute.
Yoshua Bengio did not write this page.
We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.
What Yoshua Bengio says most about
neural networks 6
- 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.
- Deep generative models have had less impact than discriminative models because of intractable probabilistic computations and difficulty using piecewise linear units.
computer vision 3
- Graph transformer networks allow multimodule document recognition systems to be trained globally with gradient methods to minimize overall performance.
- With suitable architecture, gradient-based learning can form decision surfaces that classify high-dimensional patterns such as handwritten characters with minimal preprocessing.
AGI 2
- Racing to build AGI first is a race where everyone could lose.
- Human-level AI will arrive in 5 to 20 years with 90% confidence, not in the decades-to-centuries range previously assumed.
LLMs 2
- The race between AI companies to build ever more powerful systems is a choice they made, not an inevitable law of nature, and it is a race everyone loses.
- There is currently no reliable technology to verify compliance with international AI agreements.