This lesson is all about Salesforce’s Trusted AI Principles. If you want to know Salesforce’s approach to AI and prepare for the AI Associate certification test, you need to know this. We’ll break down each principle so you can see how AI remains responsible, ethical, and aligned with human values.
Explainability vs. Transparency
We talked in previous lessons about explainability. Are explainability and the AI Trusted Principle “transparency” the same? Not quite!
Transparency: Sharing how AI systems work and what data they use.
Explainability: Making AI predictions, results, and actions understandable to non-technical users.
Example: Transparency would be including statements in a company privacy policy about how custom’s data is being used by AI and how within the organization. Explainability would be having a clickable footnote in a customer service chatbot’s response that tell the user how the AI bot decided to recommend a resolution if it used some source documents or customer information.
Why it matters: Together, they ensure AI is not only understood but trusted.
Consider This
Some thoughts to chew on about AI, Ethics, and Trusted Principles:
The term “Artificial Intelligence” was coined in 1956 at a Dartmouth Conference. Imagine what life was like in 1956. What would you say at that conference if you could go back in time now knowing how AI is progressing?
Salesforce is part of the Partnership on AI, working alongside other tech giants to promote ethical AI development. Do you think Salesforce is doing the right thing promoting ethical development?
Transparent AI is about understanding as much as it is about trust. When’s the last time you engaged with a company and they broke your trust? Are you still using their products or services?
When’s the last time you engaged with a company and they broke your trust? Are you still using their products or services? Share your thoughts with the community and see what others had to say!
Thank you! Your comments have been submitted.
The AI Fundamentals Podcast
Episode 14: Trusted AI Principles
Salesforce’s Trusted AI Principles are:
Responsible
Accountable
Transparent
Empowering
Inclusive
Each principle contributes to Salesforce’s commitment to building AI that benefits society while protecting individual rights.
Responsible
Salesforce emphasizes the responsible use of AI by:
Working with human rights experts to ensure AI respects and protects fundamental rights.
Educating and empowering customers and partners, equipping them to use AI ethically and responsibly.
Developing AI openly, sharing research findings and innovations with the wider community for the greater good.
Example in Practice: Salesforce partners with organizations like the World Economic Forum to align AI practices with global human rights standards, ensuring AI respects individuals’ autonomy.
Accountable
To ensure accountability:
Salesforce invites customer feedback to improve AI systems based on real-world experiences.
An Ethical Use Advisory Council offers guidance on AI deployments, helping Salesforce navigate complex ethical concerns.
External ethics reviews of AI research papers provide independent assessments, ensuring research remains unbiased and responsible.
Example in Practice: The external ethics review panel evaluates research papers and projects, ensuring AI systems are fair, just, and compliant with ethical standards.
Transparent
Transparency is key to building trust:
Salesforce strives for model explainability, making AI understandable to both technical and non-technical users.
Customers retain control over their data and AI models, ensuring data privacy and security are maintained.
Salesforce provides clear disclosure of AI usage terms, so customers understand exactly how AI is used in their systems.
Example in Practice: Salesforce enables customers to view and understand how AI-driven predictions, like those in Einstein, are generated, ensuring they can trust and validate the insights.
Empowering
Salesforce AI is designed to empower users:
Tools like Einstein allow users to build AI-powered applications with clicks, not code, democratizing AI development.
Salesforce offers free AI education through platforms like Trailhead, providing learning resources to all.
Salesforce delivers breakthrough AI research, advancing the state of the art in AI to benefit all users.
Example in Practice: With Einstein Prediction Builder, non-developers can create AI models to predict business outcomes without writing any code.
Inclusive
Salesforce is committed to building inclusive AI:
AI models are tested on diverse datasets to minimize bias and ensure fair outcomes across different demographics.
Salesforce holds consequence scanning workshops to evaluate the broader impact of AI on society.
Salesforce builds inclusive teams to develop AI, ensuring diverse perspectives are considered throughout the development process.
Example in Practice: Salesforce tests Einstein AI models using diverse data to ensure that recommendations work fairly for all users, regardless of their background.
To summarize:
Responsible AI means ensuring AI respects human rights.
Accountability is about taking responsibility for AI’s actions and inviting external oversight.
Transparency ensures users understand how AI works and are in control of their data.
Empowering AI means making AI accessible to everyone, even without technical expertise.
Inclusive AI ensures AI benefits everyone and minimizes bias.
If you’re internalizing these principles and keeping them in focus as you move forward, you’re contributing to a future where AI is trusted and benefits all of humanity. For more in-depth reading on Salesforce’s Trusted AI Principles, you can explore this resource: Meet Salesforce’s Trusted AI Principles.
Drop Into Focust
What does it mean for AI to be responsible?
Responsible AI ensures AI respects human rights, educates users on ethical practices, and is developed openly to benefit society.
How does Salesforce ensure accountability in AI?
Accountability involves customer feedback, guidance from an Ethical Use Advisory Council, and external reviews to ensure fair and ethical AI practices.
Why is transparency important in AI?
Transparency builds trust by making AI explainable, ensuring customers control their data, and providing clear AI usage terms.
How does Salesforce empower users with AI?
Salesforce empowers users by offering tools like Einstein, free AI education on Trailhead, and breakthrough AI research for everyone.
What makes AI inclusive?
Inclusive AI ensures fair outcomes by testing on diverse datasets, conducting consequence scanning workshops, and fostering diverse development teams.
Can you give an example of responsible AI in action?
Salesforce partners with organizations like the World Economic Forum to align AI practices with global human rights standards.
How does Salesforce ensure data privacy and control?
Salesforce ensures data privacy by giving customers control over their data and AI models and maintaining clear AI usage disclosures.
Quiz Time!
Take this quiz to test your knowledge!
Time limit: 0
Quiz Summary
0 of 3 Questions completed
Questions:
Information
You have already completed the quiz before. Hence you can not start it again.
