Article

Artificial intelligence: ethical, transparent, and reliable solutions

No items found.

The study of Artificial Intelligence potential involves multiple levels and teams within TESISQUARE. Aligned with its governance policies and Code of Ethics, the company promotes the responsible use of AI, fully aware of its opportunities and potential risks.

Artificial Intelligence offers unprecedented opportunities, yet its use demands new levels of responsibility and awareness. Its impact on individuals and organizations has raised many questions around ethics, data governance, trust, and legal principles. According to the Accenture Tech Vision 2024 report, only 35% of consumers trust how AI is implemented, while 77% believe organizations should be held accountable for any misuse of the technology.

AI provides numerous benefits for companies, which must remain aware of the actions needed to ensure its safe, responsible, and ethical development and usage. With the enactment of the AI ACT (Regulation 2024/1689, dated June 13, 2024), the European Union has established clear guidelines on the use and provision of AI systems, setting forth a series of rules and principles to follow when these systems are implemented in professional environments. Key principles include:

  • Transparency: AI system providers must deliver clear, understandable information about their systems, including capabilities, limitations, and potential risks.
  • Conformity Assessment: Providers of high-risk AI systems (e.g., for personnel management, education access, security, and justice) must have their systems assessed by independent bodies that meet the Regulation’s standards.
  • Market Surveillance: National authorities are tasked with monitoring the market to ensure AI systems comply with the Regulation and to take necessary measures.

The Regulation aims to promote the development of human-centered, trustworthy AI, while safeguarding high standards of health and safety and upholding the fundamental rights enshrined in the EU Charter of Rights.

Ethics and Responsibility

Tesisquare dedicates significant resources and energy to understanding the potential of Artificial Intelligence, both to enhance internal processes and to expand its offerings, providing clients with increasingly accurate and precise services. While the transformative potential of AI is recognized, the associated risks are equally acknowledged. Therefore, the company is committed to using AI in an ethical, responsible, and secure manner to improve efficiency, innovation, and the quality of its products and services.

Tesisquare's AI usage focuses on two key areas:

  1. Process Efficiency: AI supports decision-making and production processes by enabling faster, more precise data analysis and extraction than manual efforts. AI-powered software can aid TESISQUARE staff in handling repetitive and standardized tasks, and it also assists product development teams with source code creation.
  2. Development and Commercialization of AI-Driven Software and Algorithms: AI applications used by TESISQUARE and developed for its clients undergo continuous evaluation to monitor performance, fairness, and adherence to the company’s Policy and Code of Ethics. Any identified issues are promptly addressed and corrected. The company also invests in training personnel involved in developing and utilizing AI-based systems.

From AI Compliance to Competitive Advantage

In both the use of AI to enhance and optimize internal processes and in AI development and commercialization activities, Tesisquare adheres to the following guidelines:

  • Legal Compliance: Tesisquare is committed to abiding by AI laws and regulations, ensuring compliance with and adapting to any regulatory updates. The AI ACT adopts a risk-based approach and imposes differentiated obligations on AI system providers and developers based on risk levels: unacceptable, high, limited, and minimal or no risk. TESISQUARE is dedicated to adhering to these guidelines. Therefore, it does not develop or utilize systems with certain features, such as:
    • AI techniques that operate subliminally without user awareness
    • Systems exploiting vulnerabilities due to age or physical/mental disabilities
    • AI systems that assess or classify individuals’ reliability based on social behavior or known/predicted personality traits
    • Real-time remote biometric identification systems in public spaces
    • Biometric categorization systems using sensitive attributes, such as gender, race, ethnicity, citizenship, religion, or political orientation
  • Human-in-Command (HIC) Principle and Accountability: For Tesisquare, AI is a tool at the service of people. Users can determine when and how to leverage AI results, viewing the technology as a decision-support tool rather than a replacement. Critical decisions should always be human-led. This principle is closely tied to the concept of accountability; while AI outputs can assist decision-making, they cannot substitute human judgment. This means that AI outputs must always undergo supervision by TESISQUARE employees who work with AI applications. For example, using generative AI to write code does not relieve the developer of verifying the accuracy, integrity, and compatibility of the code. Similarly, AI-generated images or social media posts must be checked to ensure they do not infringe on third-party intellectual property rights. AI systems often function as “black boxes,” making it crucial to use their outputs with caution.

AI Usage Guidelines

TESISQUARE acknowledges that AI can significantly enhance the quality and efficiency of certain internal processes, provided that it aligns with the principles outlined. Below are additional guidelines implemented within the organization to strengthen the protection of TESISQUARE's software source code.

  • Confidentiality and Protection of Tesisquare’s Know-How and Intellectual Property: Employees are strictly prohibited from inputting Tesisquare's confidential information into publicly accessible language models.
  • GDPR Compliance and Personal Data Protection: Users are not allowed to input personal data into AI systems, especially those open to the public, ensuring compliance with data protection regulations.
  • Transparency: The use of AI systems to generate specific outputs (slides, analyses, reports, etc.) must always be disclosed to the relevant stakeholders.
  • Caution: Due to the opacity of neural network processes and the inability to access all data simultaneously processed by deep learning-based AI systems, AI is sometimes referred to as a “black box.” While AI systems may produce consistent outputs, it is not always clear how these results are achieved or which data most influence the decision-making process. Unlike traditional software, where all code is controlled and human-designed, AI systems require heightened vigilance whenever they are used to generate content, documents, analyses, and reports.

Principles for AI System Development and Commercialization for Clients

Tesisquare is committed to developing and delivering AI systems to its clients that comply with legal requirements, particularly the AI ACT, and align with guidelines established by national and international AI authorities. These systems are developed according to the following principles:

  • Responsible Use of Client Data: Tesisquare is responsible for safeguarding a large volume of client data. The use of this data for advanced analysis through AI systems is conducted transparently, adhering to confidentiality obligations and GDPR and other applicable legal requirements. Tesisquare uses client data responsibly, addressing any legal and ethical considerations as they arise.
  • Transparency: AI systems are designed and developed to ensure a level of transparency that enables users to interpret the system's output and use it effectively. Tesisquare guarantees that its AI systems are built so that individuals are aware they are interacting with an AI system unless it is obvious from the context of use.
  • Traceability and Explainability: Tesisquare develops and distributes AI systems with capabilities that allow automatic logging of events during operation. The company also provides comprehensive technical documentation describing system functionality and training processes.
  • Security: Tesisquare designs and distributes AI systems to achieve appropriate levels of accuracy, robustness, and cybersecurity, given their intended purpose. The company maintains these standards consistently throughout the entire lifecycle of the AI systems.

Read more

Related articles

Corporate
24/10/2024
Artificial Intelligence, record growth in Italy
6 out of 10 large Italian companies have launched an AI project
read article
Corporate
17/10/2024
Important updates for e-invoicing in France
Initially, the PPF, which served as the central hub of the French Electronic Invoicing and Tax Reporting model, was designed...
read article
Corporate
9/10/2024
Nuovo servizio Tesi Carrier HUB
Il nostro nuovo servizio CARRIER HUB ha l'obiettivo di semplificare e rendere più efficienti le operazioni di trasporto all'interno della...
read article

Download the brochure

Set out your requirements in the contact form and get in touch with us