How generative AI is changing investment recommendations
1st article of the serie
Artificial intelligence (AI) is no longer just a technological promise but an active part of financial market operations and communication. Generative AI tools are increasingly present in this environment, being incorporated to improve investor relations and expand access to information.
This evolution brings evident gains – such as agility, personalization, and efficiency – but it also raises uncertainties and regulatory challenges. Therefore, it is essential that the market knows how to adopt and integrate these tools and that consumers are able to use them consciously, also understanding the necessary care for safe and responsible use.
When it comes to the use of generative AI for investment recommendations, caution needs to be accompanied by study and preparation. It is necessary to analyze how regulators are dealing with this issue.
In the field of investment recommendations, the use of generative AI expands the possibilities of interaction, categorization, and personalization, allowing investors to receive information adapted to their profile and financial moment. These tools can support product analysis, simulate scenarios, and facilitate the understanding of risks and opportunities.
However, it is important to look at their performance as an instrument of support, and not a substitute for human analysis and the professional role of investment advisors and consultants, avoiding confusing technological convenience with specialized, specific and individual guidance for the investor.
In this scenario, both the Central Bank of Brazil (BCB) and the Securities and Exchange Commission (CVM) have actively positioned themselves to welcome AI, with a strategy that goes beyond regulation, involving training, governance, and institutional learning.
On the part of the BCB, a significant step was the creation of the Center of Excellence in Data Science and Artificial Intelligence (CdE IA), a community of experts dedicated to establishing guidelines for the safe and ethical use of technology within the agency itself. This movement is an important indicator for the market: before imposing rules, the regulator seeks to understand technology, interpreting its limits and potentialities.
In addition to this internal structuring effort, the BCB defined the study of the risks and impacts of the use of artificial intelligence by financial institutions as one of the priority goals for the 2025–2026 biennium. The initiative reinforces the agency's concern to evaluate, in a technical way, the effects of the adoption of these technologies on the financial system and the relationship with the consumer.
The BCB's innovation agenda also includes flagship projects such as Drex (the Digital Real) and Open Finance, which should benefit directly from the application of AI technologies. The advancement of AI is aimed at enhancing these initiatives, allowing greater automation in data analysis, improving personalized services, and increasing efficiency in verification and security procedures. With this, AI is no longer just a support instrument and becomes integrated, in a transversal way, with the modernization infrastructure of the Brazilian financial system.
CVM, in turn, has been structuring initiatives that demonstrate its direct involvement with the topic and its concern to bring regulation and innovation closer together.
The Center for Regulation and Applied Innovation (CRIA) was created precisely to establish this bridge between CVM and innovative technological solutions. It works as a space for controlled testing and dialogue with the market, allowing new tools – such as AI-based recommendation models – to be evaluated in a technical and cautious way, without stopping the advance of innovation.
In addition, CVM Resolution No. 233/2025 established the Intelligence Development Management (GDI/SDE), a unit focused on the application of data analysis and artificial intelligence technologies in the agency's supervision and inspection processes. The creation of GDI reflects CVM's commitment to understanding the use of AI in a practical way and improving its ability to monitor the market based on evidence and technological efficiency.
Overall, the message that can be drawn from the actions of regulators is that: innovation is not incompatible with protection. Technology is welcome in the financial sector, if it is accompanied by governance, transparency, and human monitoring – essential elements to maintain investor confidence and market integrity.
The proactive stance of regulators to understand the challenges arising from the adoption of technologies and ensure market security is valued by Bill No. 2338/2023 (PL), which provides for the development, promotion, and ethical and responsible use of AI in Brazil, currently pending in the Chamber of Deputies.
The Bill provides that the National System for Regulation and Governance of Artificial Intelligence (SIA), in the position of residual regulator, will be the competent authority to regulate the use of AI for economic activities in which there is no specific sectoral regulator and must collaborate with regulators of transversal themes in cases where such regulators exist.
In other words, the Bill establishes general principles, obligations, and limits (transparency, explainability, mitigation of biases, etc.), but does not intend to replace existing or future specific regulations in the regulated sectors, allowing BCB and CVM to create or apply their own regulations for the use of AI in their markets.
And this is nothing new. The EU Artificial Intelligence Act (IA ACT) - the world's first comprehensive legal framework specifically aimed at regulating the use of AI - provides that, in certain circumstances, the obligations provided for in the IA ACT may be considered fulfilled upon compliance with requirements provided for in the applicable sectoral legislation. A notable example concerns exactly financial services.
In this new context, generative AI-based tools have the potential to simplify the understanding of financial products, reduce language barriers, and democratize access to investment. For the investor, this means more information, autonomy and personalization. At the same time, it is important to recognize that these systems operate based on large volumes of data and correlation patterns, which require critical interpretation and awareness of their limits.
AI can support decision-making, but it is not yet a substitute for individual assessment or professional follow-up. The investor now has a new role: to understand the technology that helps him. And that includes understanding what it is and what it is not. As regulation evolves, the user also needs to adapt to interact with these tools safely and consciously.
After all, if the user does not know how to differentiate the methods, safeguards, and protections arising from professional and regulated advice from the simple use of AI tools, he/she may end up deviating from the desired investment route.
And the concern is not exclusive to the investor. Investment platforms also need to be aware of the independent use of AI agents in investment automation. Being a reality that imposes itself, each agent needs to define a relationship strategy with these tools. This strategy can range from intolerance to official cooperation, requiring operational risk considerations, trade-offs with other products, and effect on your business positioning.
After all, compatibility with AI agents will bring risks, many of which still have no clear regulatory treatment, but it will certainly serve as a competitive differential between investment platforms.
One thing is certain, the use of generative AI in financial recommendations and interactions holds promise loaded with a lot of potential. BCB and CVM have been showing that it is possible to keep up with innovation without renouncing stability, ethics and investor protection. The main challenge for financial institutions is to balance technology and trust – and it is precisely on this balance that the future of the relationship between artificial intelligence and investment depends.
