ETHICAL AI
ETHICAL AI
Humans are at the heart of AI
As AI becomes more far-reaching and prevalent, it’s crucial to remember who AI is meant to benefit — people. We believe ethical AI is embedded with human values and co-created by humans and machines. Sigma.AI’s responsibility to human wellbeing reaches beyond ethical training data and into our hiring practices and data privacy policies.
Humans in the loop build better AI
When humans contribute their understanding and judgment to training data, it makes for smarter AI — this is the concept behind human-in-the-loop. But AI must be trained by those who represent the people the AI is meant to serve — or else it can perpetuate biases.
We actively work to fight bias in AI through human-in-the-loop practices. From assessing the coverage and balance of training data, to selecting annotator teams that represent all groups within a population, to continuous feedback loops during the annotation process to assure fair and consistent labeling, we’re dedicated to building technology through human-machine collaboration.
500+
Languages and dialects spoken by our annotators
72%
of annotators are women, 94% of project managers are women
100+
Nationalities represented across five continents
78%
Of remote workers from rural or low-income areas
Creating opportunities with data annotation
Our human-centric culture means that we hire with respect for AI end users and our employees alike. Instead of crowdsourcing annotators, which can introduce errors and biases to AI, we rely on our own highly trained workforce of 25,000 annotators, linguists and subject matter experts working from 100+ countries across 5 continents, selected based on client project needs
We believe in creating opportunities that treat everyone with equal respect and dignity. We pay our annotators a living wage, create jobs for remote annotators working in rural, low-income and high-unemployment areas, and work with the Juan XIII foundation to engage employees with disabilities. All employees, regardless of role, are required to sign and follow our inclusion and diversity policy.
Sustainability is an everyday practice
Creating a livable world for the future takes consistent, everyday effort. Our goal is to help ensure human wellbeing in a sustainable world through our business practices and decisions. Through our sustainability policies, we strive to minimize the impact of our operations on the environment, from the resources and energy we consume to how we reuse and recycle the materials we need
- We recycle 100% of the materials we use
- 100% of energy used from renewable sources
- Energy use reduced with low-demand servers and devices
- 84,000 plastic bottles saved — we filter our tap water
Let’s work together to build smarter AI
Whether you need help sourcing and annotating training data at scale, or you need a full-fledged annotation strategy to serve your AI training needs, we can help. Get in touch for more information or to set up your proof-of-concept.