Ensuring Ethical Standards in the Development of Autonomous and Intelligent Systems

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11240%2F24%3A10481267" target="_blank" >RIV/00216208:11240/24:10481267 - isvavai.cz</a>

Výsledek na webu

<a href="https://verso.is.cuni.cz/pub/verso.fpl?fname=obd_publikace_handle&handle=VZVAhfSK3d" target="_blank" >https://verso.is.cuni.cz/pub/verso.fpl?fname=obd_publikace_handle&handle=VZVAhfSK3d</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/TAI.2024.3387403" target="_blank" >10.1109/TAI.2024.3387403</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Ensuring Ethical Standards in the Development of Autonomous and Intelligent Systems

Popis výsledku v původním jazyce

Considering ethical values, maintaining ethical requirements, and addressing ethical concerns is a crucial aspect of developing systems based on artificial intelligence (AI). Developers need to prioritize ethical considerations throughout the development process, from the initial concept to the final implementation. Despite the issuance of numerous ethical guidelines for technology developers in recent years, research results have shown no significant impact of ethical requirements on the human decision-making of software engineers from the tech community. It is therefore important to have well-defined processes to ensure that ethical standards are maintained throughout the development process. This is essential to fostering trust in AI-based systems and reducing their potential risks and negative impacts. This paper provides an overview of the processes that can integrate ethics into development methodologies by introducing a certification program that focuses on ethical accountability, transparency, privacy, and the avoidance of unacceptable algorithmic bias. The certification method specifies detailed criteria for the independent evaluation, conformity assessment, and certification of AI systems. The article also underlines factors that can affect ethical accountability, transparency, privacy, and unacceptable bias. It provides examples of the necessary documented evidence for conformity assessment against the suite of certification criteria that enable evaluation of the organizations&apos; ability to meet ethical requirements.

Název v anglickém jazyce

Ensuring Ethical Standards in the Development of Autonomous and Intelligent Systems

Popis výsledku anglicky

Considering ethical values, maintaining ethical requirements, and addressing ethical concerns is a crucial aspect of developing systems based on artificial intelligence (AI). Developers need to prioritize ethical considerations throughout the development process, from the initial concept to the final implementation. Despite the issuance of numerous ethical guidelines for technology developers in recent years, research results have shown no significant impact of ethical requirements on the human decision-making of software engineers from the tech community. It is therefore important to have well-defined processes to ensure that ethical standards are maintained throughout the development process. This is essential to fostering trust in AI-based systems and reducing their potential risks and negative impacts. This paper provides an overview of the processes that can integrate ethics into development methodologies by introducing a certification program that focuses on ethical accountability, transparency, privacy, and the avoidance of unacceptable algorithmic bias. The certification method specifies detailed criteria for the independent evaluation, conformity assessment, and certification of AI systems. The article also underlines factors that can affect ethical accountability, transparency, privacy, and unacceptable bias. It provides examples of the necessary documented evidence for conformity assessment against the suite of certification criteria that enable evaluation of the organizations&apos; ability to meet ethical requirements.

Druh

J_SC - Článek v periodiku v databázi SCOPUS

CEP obor

—

OECD FORD obor

60301 - Philosophy, History and Philosophy of science and technology

Projekt

<a href="/cs/project/VI04000107" target="_blank" >VI04000107: Krizový distribuční systém</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

IEEE transactions on artificial intelligence

ISSN

2691-4581

e-ISSN

2691-4581

Svazek periodika

5

Číslo periodika v rámci svazku

12

Stát vydavatele periodika

US - Spojené státy americké

Počet stran výsledku

10

Strana od-do

5863-5872

Kód UT WoS článku

—

EID výsledku v databázi Scopus

2-s2.0-85190172901