At Globant our mission is to make the world a better place, one step at a time. Leveraging our cross-industry expertise and Artificial Intelligence (AI) knowledge , we aim to reinvent businesses, helping organizations thrive while changing the way they make an impact on society.
The use of AI is impacting businesses and industries, bringing new challenges, and along with this , a great responsibility.To incorporate AI in the organization, we must unite ourselves and embrace a common vision, overcoming ethical challenges and social risks.
Globant has defined a set of principles that includes what we believe in, and what we encourage. We invite you to read them, embrace them and share them:
1. Augmented Intelligence: AI should exist to cooperate with humans and to improve humanity. Collaborating with humans on complex tasks and facilitating their work, relieving them from tedious tasks and elevating them.
2. Respectful Data: A data-driven culture means to have clean and accurate data, compliant with laws and regulations, guaranteeing privacy and intimacy of all individuals. Meet strict reliability, security and integrity standards.
3. Fairness: We will actively promote data-driven outcomes that are unbiased in terms of race, ethnicity, gender, nationality, income, sexual orientation, ability, and political or religious belief. We can’t expect to always have an unbiased history, so the way we use the results needs to take fairness of application into consideration as well.
4. Transparency: Create products that are transparent in their purposes and results. We are committed to pursue algorithmic accountability. AI products must preempt the risks of user data misuse and protect from imprudent use. Must exercise caution by anticipating adverse consequences. Limitations and risks should be explicit, enabling policy makers to hedge for critical cases.
5. Social Contribution: Ensure access to relevant forms of knowledge, promoting fundamental skills and critical thinking among the community. Open, promote and make more accessible AI research to the community.
6. Sustainable AI: The way to frame the problems to solve, the usage of foundational models, training strategy and other factors need to take into account how to minimize the ecological and human detrimental impact. Reduce the impact of large energy consumption on model training and serving, while improving digital sobriety and minimizing addictive behaviors.
We will not pursue any AI applications which contravene or may contravene any law or regulation, the public order and good morals, which includes, among any others:
1. Un-inform: We will not collaborate on AI systems developed to spread untrustworthy information, misinformation or disinformation. While those categories may seem contentious, appropriate sourcing of information and proper alignment of objectives can dampen unfortunate impacts.
2. Malicious Use: The same algorithm can be used in different contexts and ways. The dissemination or misuse of algorithms poses a threat to people’s capacity to trust in the use of algorithms and their legitimacy.
3. Taboo Exploitation: We need to be conscious about the misuse of sensitive characteristics such as race, ethnicity, gender, nationality, income, sexual orientation, ability, and political or religious belief. These characteristics may be leveraged for a fairer approach to AI, but their sole consideration poses concerns that need to be addressed.
4. Reckless AI: Implementations that may cause or directly facilitate injury to people. Good intentions without proper consideration can cause increased damage.
5. Law Contradiction: Projects that contravene widely accepted principles of international law and human rights, such as data protection, Information security and others.
This is an open guidance, to be revised and adapted according to the evolution of knowledge and techniques. If you are interested in learning more, email us at email@example.com.