Investor Relations
-
- pt - - -
Investor Relations
Governance · Ethics · Innovation
We develop and adopt Artificial Intelligence in an ethical, transparent, secure and people-centered way, with structured governance and alignment to international best practices.
Petrobras believes that Artificial Intelligence must serve people, respect fundamental rights and expand our ability to create value safely and sustainably.
We are committed to developing and adopting AI systems responsibly throughout their entire life cycle — from design to monitoring and decommissioning. This commitment is anchored in our Code of Ethical Conduct and our sustainability and integrity commitments.
Our Responsible AI governance is multidisciplinary and participatory, supported by governance structures operating at strategic, tactical and operational levels, ensuring that technological innovation goes hand in hand with responsibility, proportionality and precaution
Our AI activities are guided by five core principles set out in our Responsible AI Guideline, which translate into concrete commitments applied to every AI system.
We respect fundamental, individual and social rights, ethical principles and corporate values, promoting safety, fairness, diversity, equity and inclusion.
AI systems are subject to human oversight proportional to the risk level, supported by clear governance structures and formal assignment of responsibility for decisions.
We communicate the purpose, capabilities, limitations and data sources of AI systems, with a level of detail and explainability proportional to the risk and impact of the AI system.
We implement safeguards so that AI systems operate in a robust and secure manner, ensuring confidentiality, integrity, availability and the privacy of personal data.
We combine innovation with responsibility, sustainability and efficiency, guided by continuous risk and impact assessments and by the principles of precaution and proportionality.
We protect privacy and personal data in the use and development of AI, in compliance with the LGPD and our Code of Ethical Conduct, applying the Privacy by Design principle.
We apply our corporate information security controls to AI systems throughout the entire life cycle, protecting data, models and infrastructure.
We assess and mitigate undesired bias in AI systems, seeking fair and non-discriminatory outcomes.
For critical decisions, we maintain qualified human oversight, with authority to review, intervene in and reverse AI-generated outcomes.
We seek transparency about where and how we use AI and adopt levels of explainability proportional to the risk of each AI system, disclosing AI-generated content.
We use AI in the decision-making process, but responsibility remains with people and the governance chain defined in our processes.
We set explicit limits on what our AI systems can and cannot do, assessed per use case before deployment.
We consider the efficiency and environmental impact of data centers, in line with our sustainability commitments.
We do not develop or deploy AI systems that engage in manipulation, exploitation of vulnerabilities, or social scoring.
We provide channels for people affected by decisions supported by AI systems to request review and human intervention.
We promote employee training on the ethical and secure use of AI systems.
Our Responsible AI governance is multidisciplinary and participatory, organized across strategic, tactical and operational levels that connect strategy to execution and to continuous risk monitoring.
Before deployment, every AI system — developed internally or provided by third parties — undergoes a structured risk assessment process, revisited whenever the context of use changes.
We define controls and implement human oversight proportional to the risk of each AI system, with qualified authority to review, intervene and reverse AI-informed decisions. Continuous monitoring throughout the lifecycle ensures that AI systems remain aligned with our principles of responsibility, security and compliance.
Our Responsible AI governance is built in dialogue with the leading global references for AI ethics, regulation and risk management — combining international principles, regulatory frameworks and sector best practices.
Recommendation on the Ethics of Artificial Intelligence, grounded in human dignity and rights, inclusion, environmental sustainability, transparency and human oversight.
Principles for trustworthy AI: inclusive growth and well-being, transparency and explainability, robustness and security, and accountability across the life cycle.
European regulation that classifies AI systems by risk level, setting out prohibited practices, reinforced obligations for high-risk AI systems and transparency requirements.
A comprehensive, continuously updated database that catalogs and classifies AI risks, supporting structured identification, assessment and mitigation across our AI systems.
Principles for the responsible use of AI that reflect the specific challenges and opportunities of the oil and gas industry, addressing decision-making transparency, regulatory compliance, operational safety and the governance structures essential to AI adoption in the sector.
In line with our Guideline and global frameworks, we prohibit the development, acquisition or use of AI systems that violate the Code of Ethical Conduct or fundamental rights, especially those that:
