AI Policy
At Salesforce Dream Team (“SDT”), we are committed to the ethical development and deployment of Artificial Intelligence (“AI”) technologies for all. This policy outlines our approach to ensuring that our AI systems are developed and used in a manner that respects privacy, ensures fairness, maintains transparency, and promotes accountability, security, and safety for internal and external stakeholders. Our commitment extends to continually assessing the social, environmental, and legal impacts of our AI components, capturing our dedication to ethical standards and the well-being of all our stakeholders.
​
1. Data Ethics and Privacy
a. Data Collection. We commit to legally and ethically collecting data, ensuring explicit consent from individuals when required. Our data sourcing methods prioritize ethical considerations, respecting individuals’ rights and dignity above all.
b. Data Privacy. Our systems implement rigorous measures to protect personal and sensitive data, adhering to both U.S. and international privacy laws and standards, ensuring the highest levels of confidentiality and integrity possible.
c. Data Anonymization. Whenever feasible, we anonymize data to safeguard individual identities, reducing the risk of personal data exposure or PII.
d. Data Usage Transparency. We are transparent about how we use collected data, updating our communications clearly to inform users of their data’s application and use.
​
2. Fairness and Bias Mitigation
a. Bias Detection and Correction. We regularly audit our AI models for biases related to gender, race, age, and other protected characteristics and take corrective actions to mitigate identified biases. This may be performed internally or by a 3rd party.
b. Diversity in Training Data. To minimize bias, we ensure our training data encompasses diverse demographics, reflecting the variety of our global user base and candidates.
​
3. Transparency and Explainability
a. Decision Transparency. We provide clear, comprehensible explanations of our AI’s decisions and recommendations, fostering trust and understanding for all.
b. Model Explainability. Our AI models are designed to be interpretable, enabling stakeholders to grasp the processes behind decision-making we deploy.
c. Openness. We openly disclose our AI system’s capabilities and limitations, ensuring users are fully informed about the technology they interact with and its implications.
​
4. Accountability and Control
a. Human Oversight. Mechanisms for human oversight are integral to our AI decision-making processes, ensuring ethical and practical accountability at all stages of its use.
b. Audit Trails and Error Correction. We maintain detailed records of critical AI decisions and provide clear procedures for reporting and correcting errors, ensuring our systems’ reliability and integrity is maintained.
​
5. Security and Safety
a. Testing and Security Protocols. Through regular testing for vulnerabilities and prompt manipulation, and strong cybersecurity measures, we ensure our AI systems are secure and resilient against a variety of attacks and unauthorized access or use.
b. Emergency Shutdown. We provide mechanisms for the rapid deactivation of the AI in case of malfunction or security threats, ensuring safety and control. AI functions can be removed quickly and with minimal impact.
​
6. Ethical Use and Impact
a. Social and Environmental Impact. We conduct thorough assessments to understand our AI’s potential impacts on society and the environment, aiming for positive contributions and sustainable practices. AI should be a net benefit for all parties.
b. Compliance with Ethical Standards. Our AI technologies align with ethical standards and principles from leading organizations, reflecting our commitment to responsible innovation and development.
​
7. Continuous Monitoring and Improvement
a. Performance Monitoring and Stakeholder Feedback. We continuously monitor our AI’s performance and actively seek feedback from users and stakeholders, embracing opportunities for improvement. Feel free to reach out as needed.
b. Ethical Training. Ongoing ethics training for our development and oversight teams ensures a deep-seated culture of responsibility and ethical awareness.
​
8. Legal Compliance
a. Regulatory Compliance and Intellectual Property. We adhere to all applicable laws, regulations, and standards, respecting intellectual property rights and ensuring our AI’s legal integrity.
This policy serves as a foundation for SDT's commitment to ethical AI development and usage. It is a living policy subject to updates and refinements as technology and societal norms evolve over time. We invite our users, employees, and stakeholders to join us in this commitment, fostering an environment of trust, innovation, and ethical responsibility.
​
​