In addition to serious violations such as failing to report cyberattack incidents to the authorities, companies may simultaneously be punished with fines and additional sanctions, such as loss of equipment or closure of the establishment. A specialist considers the law balanced and protective of fundamental rights.
Violations of the Artificial Intelligence (AI) legislation may result in prison sentences of up to 12 years for those who implement or intentionally misuse AI systems, as well as fines that may reach Kz 750 million for individuals and Kz 1.5 billion for companies. The sanctions are included in the draft AI Law, which has been under public consultation since September of this year by the Ministry of Telecommunications, Information Technologies and Social Communication (MINTTICS).
According to the document, actions that violate privacy and the fundamental rights safeguarded by the proposal, that promote discrimination or incite hatred through algorithms, as well as those that cause serious harm to national security, public order or the democratic process, are considered crimes punishable by one to 12 years in prison. Article 76 of the proposal also criminalises facilitating crimes against the State, including espionage, terrorism or other acts that threaten national security, in addition to those that irreparably compromise the integrity of institutions.
The collection of personal information without consent or the misuse of AI for private assessments is considered a serious administrative offence, punishable by a fine of up to Kz 1.5 billion for companies and Kz 750 million for individuals.
In practice, supermarkets, banks and other commercial establishments will now be required to explain the reasons for collecting customer data and the purpose of its use. Customers, in turn, gain the right to refuse that their data be used to train AI systems or to demand that it be anonymised, ensuring the protection of their privacy.
Failure to comply with these provisions may result in infractions and fines, as set out in the proposal. This measure reinforces the principle of transparency and explainability, considered fundamental pillars of the regulation of the AI industry. These principles impose on operators the duty to provide clear information about AI-based products and services, adequately explaining their functioning, purpose and intent.
Other violations are also considered serious, such as failing to register critical AI (used in areas such as health and education) within the legal deadline, or failing to report “significant” cyberattack incidents to the authorities (see page 4).
In addition to serious administrative offences, companies may be simultaneously punished with fines and additional sanctions, such as the loss of equipment or systems used in the offence, closure of the establishment for a period of two years, deprivation of the right to subsidies or benefits granted by a public entity for two years, and suspension of registrations, licences or any authorisations granted under this law and related legislation for a period of two years, according to Article 73 of the proposal.
For Sérgio Lopes, CEO of New Cognito, a technology company part of the Mitrelli Group, interviewed by Expansão, the draft law is, in theory, “robust” and manages to balance the protection of citizens’ fundamental rights with the promotion of technological and digital innovation. “The major challenge lies in the capacity for technical oversight (standards, laboratories and experts) and coordination among regulators, namely the future AI Authority, the Data Protection Authority (APD) and INACOM. Without this machinery, there is a risk of losing proportionality in practice,” the specialist said.
However, Sérgio Lopes acknowledges that the proposal represents an effort to ensure that the advancement of Artificial Intelligence takes place without trampling democratic principles. He emphasises that the effectiveness of this regulatory framework will only be proven once it reaches the implementation phase, a decisive test to validate intentions and mechanisms.
Technological sovereignty
In a context marked by the advancement of emerging technologies such as AI, where data is the main driving force, a new form of power is gaining ground worldwide: the technological sovereignty of nations.
Source: Expansão
