On 7 May 2024, the Chilean Chamber of Deputies introduced the Artificial Intelligence (AI) Bill, making it one of the first countries in Latin America to kickstart legislative efforts in this area. The Bill seeks to regulate the design, implementation, and use of AI systems in Chile, promoting the ethical and responsible use of the technology and seeking to safeguard people’s health, safety and fundamental rights. It is based on the EU’s regulation on artificial intelligence (the EU AI Act), adopting the same risk classification as the Act and embracing the universal ethical principles contained within it.
The passage of the Bill represents a wider trend within the Latin America region, with several countries such as Brazil, Argentina and Colombia all seeking to introduce comprehensive regulatory frameworks on AI. In this article, we examine some of the concerns with Chile’s proposal and the challenges employers may face in adapting to an increasingly regulated environment. We also take a look at some of the positives, including the potential opportunities it could unlock for businesses in the country.
___
Using the right blueprint?
The first criticism of the Bill stems from its close alignment with the European regulatory model. Chile’s reality is not the same as Europe’s, and it might be difficult to apply a series of regulations designed to be implemented in territories with very different socioeconomic levels and legal systems.
Examples of this can be seen in the Bill itself. The Bill contemplates, for example, the creation of a Personal Data Protection Agency to oversee and enforce the provisions of the legislation. It would be specifically responsible for ensuring that AI systems are regularly evaluated to ensure their security and effectiveness. And yet, its powers and responsibilities are not clearly established in the Bill. This issue has been noted by the Ministry of Science, Technology, Knowledge and Innovation, raising doubts about the State’s current capacity to carry out the significant level of oversight needed under the proposed framework.
Furthermore, there is concern about the lack of existing legal frameworks in Chilean law to address sanctions or liabilities that may arise from the use of AI, whereas these may already be established in Europe. In Chile for example, there are no rules for attributing responsibility for AI errors and so questions such as ‘who will be responsible when AI causes harm?’ are unclear. Would it be the developer, the user, the system provider?
___
Other shortcomings
Added to this are the limitations of the existing Chilean Law on the Protection of Privacy in addressing some of the challenges with using AI. For example, the Law does not regulate automated data processing or algorithm-based decision-making, meaning it does not address some of the most common situations arising from the use of AI. The current version of the draft AI Bill is also silent on these issues.
Finally, it is believed by some that the Bill will introduce a framework that disincentives national innovation. This concern has arisen from the proposed introduction of a risk rating that prohibits the use of certain systems—some of which are used daily, especially in the business world. The main recommendation of experts is to therefore make these ratings more flexible and adapt them in line with the prevailing technological developments in the country. Whether this will happen or not as part of the passage of the Bill remains to be seen.
___
What are the potential benefits?
Despite these concerns, there are also reasons to be optimistic about the Bill.
It must be emphasised, first and foremost, that the Chilean proposals seek a degree of regulatory modernisation that is not currently envisioned by other jurisdictions within the region. In this respect, it has taken a proactive and somewhat pioneering stance to the area, demonstrating both a recognition of AI’s transformative potential and a commitment to managing its associated risks responsibly.
The ethical approach the proposed legislation pursues can also be viewed as positive, expressly recognising principles such as non-algorithmic discrimination, transparency, explainability, and meaningful human oversight. While the Bill might pose a challenge to employers, it also offers opportunities for the responsible development of solutions based on AI. When combined with good governance structures that help the compliant adoption of AI, this can promote its safe use whilst ensuring that businesses (and their employees) are getting the most out of the technology.
___
Takeaway for employers
As Chile moves towards a framework for AI, employers must prepare for a shift where it is no longer a discretionary tool but a regulated and auditable asset. While the Bill’s provisions remain somewhat unclear, businesses should begin adapting their internal structures to meet future compliance demands.
*Associate
