Most federal government bodies in Brazil fail to use AI efficiently
Organizations do not use artificial intelligence or are in the early stages of development, according to a government investigation.
The majority of federal government organizations in Brazil are still in the very early stages of using artificial intelligence, according to a study by the Federal Court of accounts.
The aim of the research of the Anti-Terrorism Unit was to understand the use of artificial intelligence by the federal administration, identify the risks associated with it, and inform the evaluation of the Brazilian artificial intelligence strategy.
Using a scale ranging from 1 to 4, the study found that 38% of federal government bodies in Brazil are at Level 0 of maturity in terms of the use of artificial intelligence. This means that they do not use technology. About 33.5% of organizations are at Level 1, which means that internal discussions about AI have begun but are still based on speculation.
"In those cases, a simple lack of knowledge regarding the opportunities and benefits that can be achieved using these technologies was found to be a hindrance, suggesting a gap that goes through constraints in the workforce and reaches the leaders of these organizations," he said.
The University of Texas research also found that 28.5% of all federal government bodies are at levels 2, 3, or 4 in terms of AI maturity. Most organizations within this group (17.1%) are implementing pilot projects or more detailed proof of concept. Artificial intelligence projects are produced in about 8% of organizations, while the most advanced initiatives are carried out in 3.4% of Brazilian federal government bodies.
Criticism of the National Artificial Intelligence plan
The report of the Technical Cooperation Unit also contained several criticisms regarding the Amnesty International national plan. Published in April 2021, the strategy was intended to guide actions on research, innovation, and development of artificial intelligence and related technologies to meet the country's biggest challenges and ethics.
The report concluded that the plan did not contain a set of specific, measurable, and achievable goals, which had to be achieved within a specified period. "In practice, [the plan] can be considered as guidance only, and is more open in nature," the TCU report noted.
Furthermore, the report noted that the strategy lacked an initial reference phase and that indicators and targets for performance evaluation were unclear. The report of the Technical Cooperation Unit noted that" without fundamental values about the reality to be improved, it will not be possible to develop any relevant analysis of the results achieved," adding that the strategy only provides general data on the use of artificial intelligence in Brazil, without specifying what is intended to be measured or prioritized.
In addition, the report noted flaws in the logical model provided for in the national plan for artificial intelligence and how the plan affects identified problems or priority opportunities, as there is no discrimination in inputs, activities, products, impacts, and outcomes.
The publication of the AI strategy was preceded by a process that has lasted more than a year since the consultation was launched to gather input for the plan in late 2019, after a period of engagement with AI consulting firms and an international measurement process. According to the Brazilian government, the consultation lasted until March 2020 and more than 1,000 contributions were received.
According to the then Minister of Science, Technology, and innovation, Marcos Pontis, the publication was a "dream realization" and a big step forward for Brazil. Pontes said the Brazilian government considers artificial intelligence "essential" to the development of many other technologies, such as innovations around the IoT approach.
Comments
Post a Comment