Sigma AI is deeply committed to delivering ethical artificial intelligence projects — including generative AI — in a manner that is transparent and socially beneficial.
From annotation accuracy to bias mitigation, the ethical treatment of our people, and the practices guiding our partnerships — these are the four pillars of how Sigma builds AI you can trust.
We implement rigorous quality control measures and continuous monitoring to align data quality with human judgment from domain experts for generative AI. We can achieve up to 99.99% accuracy for traditional AI projects with objective measurements.
We offer clear documentation of our data annotation processes, making methodologies accessible to all stakeholders. This is particularly important when supporting subjective generative AI projects.
We ensure compliance with ethical standards and regulations, continually updating our practices to reflect the latest advancements in the field.
We hire multicultural staff from more than 120 countries and require bias training for our team. We provide ongoing monitoring and targeted continuing education to prevent bias.
Bias exists in the data used to train ML models, often due to historical patterns of discrimination or lack of representative data. Sigma AI identifies and reduces biases through detection in program management and tooling to pre-process data for quality, diversity, and coverage.
We advise clients on methodologies to select balanced and representative data sets or synthetic data to ensure it reflects diverse populations and scenarios.
We contribute to a better world by creating jobs, paying living wages, and providing employment and career advancement opportunities for marginalized workers. This helps reduce poverty, promote equal opportunities, empower women, and provide income for individuals in rural, remote, or low-income communities, as well as for people with disabilities through our work-from-home model.
Our staff hail from 120 countries and speak more than 600 languages and dialects. More than 70% of our project and program managers are women, and the majority of our annotators are women.
Sigma AI provides comprehensive training and continuous feedback to annotators to achieve unmatched quality. We require 100% compliance on security and safety training.
We recognize AI automation can be a challenge to some workers. We invest in education to support career advancement, upskilling them to adapt and succeed in more complex AI-related jobs.
We help our employees prevent workplace injuries and minimize fatigue with mandatory break times, training on proper ergonomics, and ergonomic devices in our workplace.
At Sigma, our focus is on people: those we work with and those we serve. Since our founding in 2008, we have upheld a Code of Ethics as a core part of all our contracts with employees, contractors, partners, and providers — reflecting our values of inclusion, equal opportunities, privacy, confidentiality, respect, empathy, and fair treatment.
We partner with organizations that share our commitment to advancing ethical AI standards across the industry.
To oversee our ethical practices, we have established robust accountability structures, including regular audits and independent reviews.
We prioritize the privacy of individuals by giving them the choice to consent to their data being used in AI. We comply with GDPR and all relevant local and national privacy laws.
Ethical AI aims to do no harm and continuously assess the impact of AI on society, while responsible AI involves putting principles into practice to create safe, fair and trustworthy AI systems. Sigma AI is dedicated to ethical AI development, transparency, and social responsibility — building AI systems that are not only advanced but also fair and beneficial to individuals and society.
Read about our approach to corporate environmental and social responsibility.
Get in touch →