www.techtarget.com Open in urlscan Pro
104.18.19.71  Public Scan

Form analysis
1 forms found in the DOM

POST https://www.techtarget.com/search

<form action="https://www.techtarget.com/search" method="post" class="header-search">
  <label for="header-search-input" class="visuallyhidden">Search the TechTarget Network</label>
  <input class="header-search-input ui-autocomplete-input" id="header-search-input" autocomplete="off" type="text" placeholder="Search the TechTarget Network">
  <button aria-label="Search" class="header-search-submit"><i class="icon" data-icon="g"></i></button>
</form>

Text Content

Enterprise AI
Search the TechTarget Network
Login Register
Explore the Network
 * TechTarget Network
 * Business Analytics
 * CIO
 * Data Management
 * ERP

 * Enterprise AI
 * * AI Business Strategies
   * AI Careers
   * AI Infrastructure
   * AI Platforms
   * AI Technologies
   * More Topics
      * Applications of AI
      * ML Platforms
   Other Content
    * News
    * Features
    * Tips
    * Webinars
    * 2023 IT Salary Survey Results
    * Sponsored Sites
    * More
       * Answers
       * Conference Guides
       * Definitions
       * Opinions
       * Podcasts
       * Quizzes
       * Tech Accelerators
       * Tutorials
       * Videos

 * Follow:
 * 
 * 
 * 
 * 
 * 


 * Home
 * AI business strategies

Rawpixel.com - stock.adobe.com

Rawpixel.com - stock.adobe.com

Feature


RESPONSIBLE AI VS. ETHICAL AI: WHAT'S THE DIFFERENCE?




ETHICAL AI ESTABLISHES PRINCIPLES FOR AI DEVELOPMENT AND USE, WHILE RESPONSIBLE
AI ENSURES THEY'RE IMPLEMENTED IN PRACTICE. LEARN HOW THE TWO DIFFER AND
COMPLEMENT EACH OTHER.

 * Share this item with your network:
 * 
 * 
 * 
 * 
 * 

 * 
 * 
 *  * 
    * 
    * 
    * 


By
 * Kashyap Kompella

Published: 27 Sep 2024

Responsible AI and ethical AI are closely related, and each offers distinct but
overlapping principles for AI development and use. Successful organizations
cannot have one without the other.



Responsible AI focuses on accountability, transparency and compliance with
regulations, while ethical AI -- sometimes referred to as AI ethics --
emphasizes broader moral principles like fairness, privacy and societal impact.
Recently, discussions about their importance have intensified, pushing
organizations to consider the nuances and benefits of integrating both
frameworks.

Responsible AI and ethical AI work hand in hand. The loftiest ethical AI
ambitions can amount to nothing without practical implementation; likewise,
responsible AI needs to be grounded in clear and purposeful ethical principles.
And AI ethics concerns often inform the regulatory frameworks that responsible
AI initiatives must adhere to, highlighting their mutual influence.

By combining both approaches, organizations can build and deploy AI systems in
ways that are not only legally sound, but also aligned with human values and
designed to minimize harm.


THE NEED FOR ETHICAL AI

Ethical AI refers to the values and moral expectations governing AI use. These
principles can evolve and vary: What is acceptable today might not be tomorrow,
and ethical standards can differ by culture and country. However, many ethical
principles, such as fairness, transparency and avoidance of harm, tend to be
consistent across regions and over time.

Hundreds, if not thousands, of organizations have expressed interest in ethical
AI by developing ethical frameworks, an important first step. AI and automation
technologies can fundamentally alter existing relationships and dynamics among
stakeholders, possibly requiring an update to the social contract -- that
implicit agreement on how society should function.

Ethical AI informs and drives these discussions, helping define the contours of
an AI social contract by establishing what is and isn't acceptable. AI ethics
frameworks often serve as a precursor to AI regulation, although some regulation
is emerging alongside or even ahead of formal ethical frameworks.

This evolution requires input from multiple stakeholders, including consumers,
citizens, activists, academics, researchers, employers, technologists, lawmakers
and regulators. Existing power dynamics can also influence whose voices shape
the ethical AI landscape, with certain groups having more influence than others.

AI ethics targets a variety of moral concerns and dilemmas related to AI.



ETHICAL AI VS. RESPONSIBLE AI

Ethical AI is aspirational, focusing on AI's long-term effects and societal
impact. Numerous ethical concerns around AI have surfaced in recent years,
especially following the rise of generative AI.

One important issue is machine learning bias, which occurs when AI systems
produce biased, stereotyped or harmful outputs due to flawed, unrepresentative
or biased training data sets and model designs. This bias is particularly
dangerous for high-stakes use cases such as loan approvals and police
surveillance, where biased output and decision-making can cause serious harm and
perpetuate existing inequalities.

Other ethical concerns include AI hallucinations, where systems generate false
information, and generative AI deepfakes, which can be used to spread
disinformation. The common thread of these ethical AI issues is that they all
threaten basic human values, such as safety, dignity, equality and democracy.

Responsible AI, in contrast, addresses both ethical concerns and business risks
-- issues like data protection, security, transparency and regulatory
compliance. It provides concrete ways to operationalize ethical AI aspirations
as responsible AI practices for each phase of the AI lifecycle, from design and
development to monitoring and usage.

The relationship between ethical AI and responsible AI is like the relationship
between a company's vision and the operational playbooks used to achieve it.
Ethical AI provides the high-level principles, while responsible AI shows how to
implement those principles in practice throughout the AI lifecycle.

Implementing responsible AI means considering several responsible AI principles.



THE CHALLENGES OF PUTTING PRINCIPLES INTO PRACTICE

Modern enterprises rely on codified business processes and practices. To be
sure, there is room for human discretion, but standardized processes are the
norm to ensure efficiency, consistency and scale. This applies to software
development, including AI, where following standard methodologies and processes
leads to many organizational benefits.

Although ethical AI can sometimes be treated as a separate initiative focused on
broader societal impacts, ethical principles are frequently included in
responsible AI frameworks. To implement these principles, organizations must
integrate them into existing processes, routines and development practices. This
is often done through user-friendly checklists, standardized methodologies,
reusable templates and evaluation guides. For this reason, AI ethics are often
included within a comprehensive responsible AI checklist.


IMPLEMENTING RESPONSIBLE AI

While ethical AI is top of mind for many organizations, it is usually embedded
into responsible AI practices. Organizations should focus on the following areas
when implementing responsible AI:

 * Transparency. Both technical and nontechnical measures can increase
   transparency. Explainable AI techniques can help make models more
   transparent, although not all complex AI systems can be fully explained. In
   addition to comprehensive technical documentation, transparency includes
   clearly communicating with users about system limitations, biases and
   appropriate use.
 * Stakeholder involvement. Responsible AI requires input from multiple
   stakeholders in the organization. These might include technical teams, legal
   and compliance, quality assurance, risk management, privacy and security,
   data governance, procurement and vendor management. Some implementations
   might also call for insights from subject matter experts in areas such as
   finance, HR, operations and marketing.
 * Documentation. A RACI matrix -- Responsible, Accountable, Consulted, Informed
   -- should outline the roles and responsibilities of each stakeholder
   throughout the AI lifecycle. To equip stakeholders to effectively contribute,
   organizations must create templates, checklists and other tools for each
   function involved.
 * Regulation and compliance. Organizations must stay agile and up to date as AI
   regulations evolve globally. In addition to complying with the EU AI Act,
   businesses should pay attention to emerging regulatory frameworks in other
   regions, such as the U.S. and China, as well as relevant local or state
   regulations. Both internal and third-party audits can help organizations
   assess and validate their compliance.
 * Third-party tools. Some organizations build their own AI systems, while
   others use third-party AI applications; many do a mix of both. Organizations
   must develop and enforce guidelines and requirements for AI procurement from
   external suppliers, specifying vendor obligations and system compliance.

Kashyap Kompella is an industry analyst, author, educator and AI adviser to
leading companies and startups across the U.S., Europe and the Asia-Pacific
region. Currently, he is the CEO of RPA2AI Research, a global technology
industry analyst firm.



RELATED RESOURCES

 * AI business strategies for successful transformation –Webinar
 * Redesigning Productivity in the Age of Cognitive Acceleration –Webinar

DIG DEEPER ON AI BUSINESS STRATEGIES

 * AI AND GDPR: HOW IS AI BEING REGULATED?
   
   
   By: George Lawton

 * EXPLAINABLE AI
   
   
   By: Alexander Gillis

 * AMA PUBLISHES GUIDANCE ON AI DEVELOPMENT, DEPLOYMENT, USE
   
   
   By: Veronica Salib

 * AMA ESTABLISHES NEW PRINCIPLES FOR AI DEVELOPMENT, DEPLOYMENT AND USE
   
   
   By: Shania Kennedy

Sponsored News
 * Sustainable AI in Action: 3 Real-World Examples –Equinix
 * How AI and the Data Center Streamlines Business Processes –HPE & Intel
 * Transform Claims Management: Enhancing Accuracy, Compliance and Efficiency
   –Zelis Healthcare
 * See More

Related Content
 * responsible AI – Enterprise AI
 * How AI ethics is coming to the fore with generative AI – ComputerWeekly.com
 * How AI ethics is the cornerstone of governance – Enterprise AI


Latest TechTarget resources
 * Business Analytics
 * CIO
 * Data Management
 * ERP

Business Analytics
 * ThoughtSpot taps Salesforce exec Karkhanis to be new CEO
   
   As analytics enters a new era dominated by GenAI, the vendor has named former
   Salesforce Sales Cloud and Einstein Analytics ...

 * 7 skills for improved data visualizations
   
   Great data visualizations require a combination of analytics, design and
   communication. Master seven key skills for data ...

 * Teradata ClearScape Analytics update targets ROI on AI, ML
   
   The vendor's latest platform update adds features such as a PySpark
   conversion tool and AutoML aimed at helping customers improve...

CIO
 * 8 KPIs to reduce technical debt
   
   To ensure an IT project's long-term success, stakeholders should keep track
   of certain KPIs to ensure technical debt is as low as...

 * Proposed ban on connected vehicle tech from China too broad
   
   The U.S. Commerce Department's proposal to ban software and hardware for
   connected vehicles imported from China or Russia might ...

 * Fears grow about big tech guiding U.S. AI policy
   
   Apple joined a list of big tech companies voluntarily committing to
   responsible AI. But without regulation, there is no legal ...

Data Management
 * Cockroach Labs adds vector search, updates pricing options
   
   The database vendor's new update includes an integration with pgvector that
   provides users with the vector search capabilities ...

 * New Grafana Labs features aim to simplify observability
   
   The vendor's latest set of capabilities includes prebuilt applications and
   automated features aimed at making it easier to use ...

 * Metadata management tools: 9 top options to fulfill key needs
   
   Metadata management tools can range from comprehensive packages of many
   features to masters of a specific niche. Consider 9 of ...

ERP
 * Plex adds connected worker capabilities to MES
   
   Connected worker capabilities are now part of the Plex smart manufacturing
   platform, designed for manufacturers to address worker...

 * AI and sustainability on tap for Oracle Cloud applications
   
   Oracle unveils new capabilities for AI use cases, smart manufacturing and
   sustainability for Oracle Cloud SCM applications at ...

 * 5 steps of an ERP implementation communication plan
   
   Communicating effectively about an ERP implementation is a must because of
   the complexity and stress of the project. Learn what ...

 * About Us
 * Editorial Ethics Policy
 * Meet The Editors
 * Contact Us
 * Advertisers
 * Partner with Us
 * Media Kit
 * Corporate Site

 * Contributors
 * Reprints
 * Answers
 * Definitions
 * E-Products
 * Events
 * Features

 * Guides
 * Opinions
 * Photo Stories
 * Quizzes
 * Tips
 * Tutorials
 * Videos

All Rights Reserved, Copyright 2018 - 2024, TechTarget

Privacy Policy
Cookie Preferences
Cookie Preferences
Do Not Sell or Share My Personal Information


Close