Submission to the Commission of Jurists on the Brazilian Artificial Intelligence Bill

PI’s submission focussed on highlighting the potential harms associated with the use of AI within schools and the additional safeguards and precautions that should be taken when implementing AI in educational technology.

Key advocacy points

Among others, PI recommended:

  • Any AI legislation must pay particular attention to the relationship between public-private partnerships. Private actors should ensure human rights-based approach to their practices, therefore the AI bill should reference full abidance to the Abidjan Principles, in particular the adoption of rules and regulations for the private sector in this area, and to the UN Guiding Principles on Business and Human Rights.

  • Companies providing AI technologies to schools should be required to waive commercial confidentiality and make their technologies fully auditable by any third party.

  • If details of the workings of a particular technology cannot be disclosed for specified and valid grounds of serious commercial harm to the company, an independent oversight body bound by duties of confidentiality should be granted full access to all details of the technology required to establish those details.

  • It should also be noted that all forms of redress codified in this legislation, should not be overly burdensome to allow for children and their families to seek redress.

Advocacy

We submitted a report to the Commission of Jurists on the Brazilian Artificial Intelligence Bill focussed on highlighting the potential harms associated with the use of AI within schools and the additional safeguards and precautions that should be taken when implementing AI in educational technology.

The use of AI in education technology and schools has the potential to interfere with the child’s right to education and the right to privacy which are upheld by international human rights standards that Brazil has ratified. Our submission argued that any legilsation introduced in Brazil to regulate AI must include appropriate protections for children who's schools may use AI.