Trustworthy AI

We believe AI should respect privacy and data protection regulations, operate in a secure and safe way, function in a transparent and accountable manner, and avoid unwanted biases and discrimination. We are committed to safe and trustworthy AI, in line with the White House Voluntary Commitments and other global AI Safety initiatives.

Our Guiding Principles for Trustworthy AI

Privacy

AI should comply with privacy laws and regulations, and meet societal norms for personal data and information privacy.

Safety and Security

Ensure that AI systems perform as intended and avoid unintended harm and malicious threats.

Transparency

Make AI technology understandable to people. Explain, in non-technical language, how an AI system arrived at its output.

Nondiscrimination

Minimize bias in our AI systems and give all groups an equal opportunity to benefit from AI.

Methods and Technologies

Trustworthy AI principles are foundational to our end-to-end development and essential for the technical excellence that enables partners, customers, and developers to do their best work. We are building data factories for generative AI services, tools to curate and validate unbiased datasets for computer vision, libraries for scaling natural language processing data for large language models, groundbreaking security features like confidential computing, and innovations like open-source techniques for model alignment with human feedback.

Our Trustworthy AI Solutions

Model Card++

An AI model card is a document that provides detailed information about how machine learning models work, encouraging transparency and trustworthiness.

NVIDIA Omniverse Replicator

NVIDIA is reducing unwanted bias and protecting privacy by generating diverse synthetic datasets to replicate real-world use cases in autonomous vehicles, industrial inspection, and robotics simulation.

NeMo Guardrails

NVIDIA NeMo Guardrails helps ensure that smart applications powered by large language models (LLM) are accurate, appropriate, on topic, and secure.

Partnering for Trustworthy AI Technology

Te Hiku Media

NVIDIA Inception member Te Hiku Media created a highly accurate bilingual speech recognition system for the Māori language and NZ English, built and owned by its own language community.

Getty Images

Generative AI by Getty Images is trained with NVIDIA Picasso on responsible, clean, and fully licensed data, opening countless opportunities for creative endeavors.

Adobe Firefly

Using NVIDIA GPUs, Adobe Firefly is Adobe’s family of creative generative AI models, designed to generate content safe for commercial use while significantly improving creative workflows.

Trustworthy AI in the News

View All Trustworthy AI Stories

A Commitment to Research

Our AI research is focused on developing algorithms and systems that can augment human capabilities, solve complex problems, and improve efficiencies across industries. We work to maintain our guiding principles of privacy, transparency, nondiscrimination, and safety and security in all research practices and methodologies.

Featured publications and papers:

Discover More at the GTC AI Conference

Hear from the experts and government leaders paving the way for AI regulation and trustworthiness. Explore NVIDIA GTC Trustworthy AI sessions curated to help companies identify and address potential obstacles to developing their own initiatives.