Dario Amodei
Co-founded Anthropic in 2021 after leading safety research at OpenAI, betting that safety and capability research need to happen inside the same lab building frontier models.
Dario Amodei led safety-focused research at OpenAI, including early work on scaling lawsScaling LawsScaling laws are empirical relationships between a model's loss and its parameter count, dataset size, and compute budget, used to plan how large to train a new model. — the empirical finding that a model's performance improves predictably as a smooth function of parameters, data, and compute, which became a central argument for why scale itself was worth pursuing aggressively. In 2021, alongside his sister Daniela AmodeiDaniela AmodeiCo-founded Anthropic and serves as its president, running the operational and policy side of a lab built around aligning frontier AI development with safety research. and several other OpenAI researchers, he co-founded Anthropic, built around the premise that AIAI (Artificial Intelligence)AI is the field of building systems that perform tasks normally requiring human intelligence — reasoning, perception, language, and decision-making. safety research is most effective when done by a team also building and shipping frontier models, rather than studied at arm's length from labs racing ahead on capability alone.
As Anthropic's CEO, Amodei has pushed RLHFRLHF (Reinforcement Learning from Human Feedback)RLHF trains an LLM to match human preferences by learning a reward model from ranked response comparisons, then optimizing the LLM against that reward.-style alignmentAlignmentAlignment is whether a trained model's actual objective and behavior match what its designers intended, split into outer and inner alignment. techniques and interpretabilityInterpretabilityInterpretability is the practice of identifying what computation a trained network is actually performing internally, instead of inferring intent purely from its outputs. research (trying to understand what's actually happening inside a trained model, not just how it behaves at the input-output level) as core research priorities alongside model capability, developing the Claude family of models. He has written extensively and publicly about both the transformative potential of powerful AI and the risks of getting safety wrong at frontier scale, positioning Anthropic's mission explicitly around that tension rather than treating it as a side concern.
See also: Daniela AmodeiDaniela AmodeiCo-founded Anthropic and serves as its president, running the operational and policy side of a lab built around aligning frontier AI development with safety research.
Learn more: AI Safety & Alignment · Wikipedia: Dario Amodei
Andrew Ng
Co-founded Google Brain and Coursera, and taught the machine learning course that introduced millions of people outside academia to the field.
Daniela Amodei
Co-founded Anthropic and serves as its president, running the operational and policy side of a lab built around aligning frontier AI development with safety research.