RAI Basic

Uploaded by

Aravind AG

Available Formats

Download as DOCX, PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

27 views

RAI Basic

Uploaded by

Aravind AG

Available Formats

Download as DOCX, PDF, TXT or read online on Scribd

You are on page 1/ 2

Responsible AI refers to the ethical and transparent development, deployment, and use of

artificial intelligence technologies. It emphasizes principles like fairness, accountability,

transparency, and inclusivity to ensure AI systems are trustworthy and beneficial for all
stakeholders
Key aspects of responsible AI include:
 Fairness: Ensuring AI systems treat all users equitably and avoid biases.
 Transparency: Making AI processes and decisions understandable and clear.
 Accountability: Holding developers and organizations responsible for AI outcomes.
 Privacy and Security: Protecting user data and ensuring secure AI operations.
 These principles help mitigate risks and maximize the positive impact of AI on society

Several leading IT companies are actively working on responsible AI initiatives. Here are some
notable ones:
 IBM: IBM has been a pioneer in promoting trustworthy AI, focusing on transparency,
fairness, and accountability12.
 Microsoft: Microsoft aims to create AI that is ethical and responsible, with a strong
emphasis on eliminating biases and ensuring transparency34.
 Google: Google, including its DeepMind division, is dedicated to ethical AI practices,
focusing on human-centered design and reducing biases34.
 Meta: Meta (formerly Facebook) is also committed to responsible AI, working on
transparency and fairness in its AI systems3.
 Accenture: Accenture designs and deploys responsible AI solutions that are ethical,
transparent, and trustworthy3.
 Deloitte: Deloitte offers an AI Risk Management Framework to help organizations govern
AI responsibly3.
 Amazon: Amazon is part of the Partnership for Global Inclusivity on AI, focusing on
responsible AI practices4

Infosys has developed a comprehensive Responsible AI framework called the Infosys Topaz
Responsible AI Suite. This suite is designed to help enterprises balance innovation with ethical
considerations, such as bias prevention, privacy protection, and compliance with regulations12.

The framework is built around three key pillars:

Scan: This involves continuous monitoring and assessment of AI systems to identify risks,
vulnerabilities, and compliance status. It includes tools like the Responsible AI Watchtower for
tracking external events and the Responsible AI Control Center for internal monitoring12.
Shield: This pillar focuses on protecting AI models from threats and ensuring they operate
securely and ethically. It includes solutions for bias detection, privacy preservation, and security
measures12.
Steer: This involves guiding AI development and deployment to align with ethical standards and
business goals. It includes governance frameworks, ethical guidelines, and compliance
strategies12.
Infosys’s approach aims to ensure that AI technologies are developed and used responsibly,
maximizing their positive impact while minimizing potential risks.

ISO (the International Organization for Standardization) and IEC (the International
Electrotechnical Commission) ISO/IEC 42001 is a management system standard that
provides requirements for establishing, implementing, and improving AI management systems.
It's applicable to all types of companies and organizations, including public sector agencies,
non-profits, and companies of any size.

What it does
ISO/IEC 42001 helps organizations ensure that AI systems are developed and used responsibly
and ethically. It does this by:
 Promoting transparency, accountability, and trustworthiness
 Emphasizing ethical principles, such as fairness, non-discrimination, and privacy
 Helping organizations identify and mitigate risks
 Encouraging organizations to prioritize human well-being, safety, and user experience
 Helping organizations comply with relevant laws, regulations, and data protection
requirements
Benefits
 ISO/IEC 42001 certification can help organizations build trust with stakeholders, gain a
competitive advantage, and make strategic decisions. It can also help organizations
improve the quality, security, traceability, transparency, and reliability of their AI
applications.
Certification process
 The certification process includes an information questionnaire, technical-economic
offer, on-site and off-site audits, and a final report. A Technical Committee evaluates the
process, and if successful, a certificate of conformity is issued. Organizations must also
undergo an annual audit to maintain their certification.