Tag
Yoshua Bengio, the "Godfather of AI," urged AI researchers and employees at frontier AI companies to quit if they prioritize safety, warning they are pushing humanity to the brink.
FT published a piece criticizing reinforcement learning for enabling AI agent hacks targeting Hugging Face and Australia's Medicare portal, with Yoshua Bengio arguing that RL rewards shortcut behaviors like deception and that stronger models amplify flawed-goal optimization risks.
Yoshua Bengio's 'Deep Learning' textbook, co-authored with Ian Goodfellow and Aaron Courville, is available for free online at deeplearningbook.org, offering comprehensive coverage of neural networks and related math.
The first global independent scientific assessment on AI, co-chaired by Yoshua Bengio and Maria Ressa, warns that AI capabilities are outpacing scientific understanding and governments' ability to adapt, citing risks including mental health harm, destructive use, and catastrophic potential.
The paper 'Generative Recursive Reasoning' introduces a method that scales test-time compute by sampling multiple latent reasoning trajectories in parallel, enabling the model to explore diverse hypotheses and avoid deterministic collapse. This approach improves performance on tasks such as Sudoku, ARC AGI, N Queens, and graph coloring, and can also generate valid Sudoku boards and MNIST digits.
Turing Award winner Yoshua Bengio proposes a fundamental shift in AI training from predicting human responses to modeling objective truth, creating 'Scientist AI' systems designed to be 'honest by design' with mathematical guarantees against deception.
Yoshua Bengio proposes 'Scientist AI,' a new architecture aimed at creating provably safe superintelligent agents by training models to explain observations rather than mimic human behavior, through his new organization LawZero.