The growing pervasiveness of artificial intelligence in financial markets is exposing retail investors to unprecedented forms of cognitive and informational asymmetry, as recently underscored by ESMA’s 2025 Warning on the use of AI-driven tools in investment decisions. The purpose of this article is to investigate the regulatory and dogmatic implications of this phenomenon at the intersection of the EU AI Act (Regulation 2024/1689), the MiFID framework, and the Italian Consolidated Law on Finance (TUF). The methodology adopted is dogmatic and comparative-regulatory: the notion of “AI system” is reconstructed as an organisational rather than a merely technological entity, and is then tested against the allocation of liability along the AI value chain (Article 25 AIA), the conceptual distinction between control and authority underpinning the deployer category, and the emerging paradigm of AI-as-a-service in financial intermediation. Particular attention is devoted to fin-influencers and large diffusion models (LDMs) as new vectors of systemic risk, and to the IOSCO 2025 Report’s recommendations on supervisory convergence. The article concludes that, where algorithmic opacity erodes the traditional pillars of investor protection, financial education must be reconceived as a constitutional precondition for the effective exercise of the economic freedoms; the principle of technological neutrality, far from being a passive regulatory stance, is reinterpreted as an active duty to safeguard the human dimension of financial choice — the actus humanus — against the seductive promises of algorithmic infallibility, which the article evokes through the metaphor of the markets’ “golden calf”. Read in this key, the contribution intersects the agenda of sustainable finance: responsible AI practices, algorithmic accountability and a renewed investor awareness emerge as preconditions for the long-term resilience of capital markets and for the integration of ESG-oriented decision-making into retail investment behaviour.
The Demands of Financial Education in the Face of the Markets' "Golden Calf": The Role of Artificial Intelligence
Attilio Altieri
2026-01-01
Abstract
The growing pervasiveness of artificial intelligence in financial markets is exposing retail investors to unprecedented forms of cognitive and informational asymmetry, as recently underscored by ESMA’s 2025 Warning on the use of AI-driven tools in investment decisions. The purpose of this article is to investigate the regulatory and dogmatic implications of this phenomenon at the intersection of the EU AI Act (Regulation 2024/1689), the MiFID framework, and the Italian Consolidated Law on Finance (TUF). The methodology adopted is dogmatic and comparative-regulatory: the notion of “AI system” is reconstructed as an organisational rather than a merely technological entity, and is then tested against the allocation of liability along the AI value chain (Article 25 AIA), the conceptual distinction between control and authority underpinning the deployer category, and the emerging paradigm of AI-as-a-service in financial intermediation. Particular attention is devoted to fin-influencers and large diffusion models (LDMs) as new vectors of systemic risk, and to the IOSCO 2025 Report’s recommendations on supervisory convergence. The article concludes that, where algorithmic opacity erodes the traditional pillars of investor protection, financial education must be reconceived as a constitutional precondition for the effective exercise of the economic freedoms; the principle of technological neutrality, far from being a passive regulatory stance, is reinterpreted as an active duty to safeguard the human dimension of financial choice — the actus humanus — against the seductive promises of algorithmic infallibility, which the article evokes through the metaphor of the markets’ “golden calf”. Read in this key, the contribution intersects the agenda of sustainable finance: responsible AI practices, algorithmic accountability and a renewed investor awareness emerge as preconditions for the long-term resilience of capital markets and for the integration of ESG-oriented decision-making into retail investment behaviour.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
