Promoted the thematic session on the second day of the 2022 edition of the Magna Meeting of the Brazilian Academy of Sciences “Information and Technology (Its Impact on Society)”. By academic Jose Roberto PoissonsJosara Marques de Almeida (UFMG) speakers, Louis sheep meat (UFRGS) and Henrique Malvar (Microsoft Research Maybe CollectionDiscuss the dissemination of false information, the importance of innovation to a country’s development and how new technologies can provide greater access to minority groups.
Online Misinformation: Fighting Our Demons
According to Almeida Brave, an affiliate of ABC between 2011 and 2015, the first reports of the spread of misinformation are much older than expected. According to a UFMG professor, the first known case of disinformation occurred in 44 BC, when Octavian launched a campaign to discredit Mark Antony, using coins as a strategy. The coins contained the image of Marc Anthony and shameful phrases. “Although disinformation is supported today by technology, it is essentially a human phenomenon,” she comments.
The impact of disinformation follows the development of human communication and its technologies. From carrier pigeons (one-on-one model), through television and radio (model one for many, or “one for one”); And currently, with social media, especially messaging apps like Whatsapp, the paradigm prevails. “A lot to many” Or many-to-many, as many keep misinformation and enter it into cycles that keep spreading.
The researcher explains that misinformation is currently being generated on the fringes of the web, first through anonymous media and expanding to Whatsapp, YouTube, Reddit, Twitter, Facebook and finally professional media. It can come in many formats: spam, photos, rumors, and even sensational articles.
“Understanding what this phenomenon is, how it occurs, and who are the actors behind it, is what motivates me,” comments the UFMG professor. “I seek to understand the effects of amplification, which include robots [robôs que ajudam na disseminação de informação falsa] and echo chambers [plataformas que auxiliam na propagação de conteúdo]. “
According to the researcher, there are certain cognitive areas in which humans cannot easily distinguish misinformation, using criteria such as the reliability factor: many close people believe that it or the content is in many places.
To get a comprehensive view of this phenomenon, in 2018 – a year marked by truck drivers’ strikes and the presidential election – a research group for the scientist began analyzing the type of content that was being shared in disinformation groups on Whatsapp, and for that, joined public access groups to collect data. The conclusions of the work show that the main type of shared content is images, mainly satirical and opinion content, which are repeated many times in other groups, blogs and Twitter. The most disturbing fact is that in a group of 200 people, a single piece of content can reach thousands of people, and even gain space on Google Images. “There is no clear link from one group to another, there are no formal paths. But there is a basic network architecture that connects groups to regular users.”
However, these networks are very noisy – and it is still difficult to determine which transmission pattern to follow. The fact that members not only work together, but also in isolation makes it increasingly difficult to know the backbone of these groups.
“Although there is no solution, my goal in this speech is to raise awareness, because I believe that is part of the solution. We need the support of the community and users – it can be a technically effective solution, but it goes beyond that,” Almeida said, noting Three solutions to the caseFact Check Scalable, education and awareness. “We need to try to restore an old concept that is a fact and not just a story.
The role of deep technology in economic development in the twenty-first century
“We are migrating into a new era,” says Louis Lamb, a professor in the Institute of Informatics at UFRGS. “It is a time of great change and a large part of society may not adapt.”
In his speech, the researcher reflected on the current moment that Brazil faces to enter the digital market, in the face of innovations: “Technology has a role in influencing human life in all aspects – the bad reputation of AI can provide improvements in agribusiness, in scientific methodologies and even in the way with which a person deals with others.
Their analyzes indicate that the world is moving towards an era in which the human ability to generate income and innovate is more important than ever – and this is proven in their list of the top 10 companies that generate the most wealth today. Names such as Apple, Microsoft, Amazon and Tesla appear on the list that in 2007 was still strongly associated with the industrial age (mainly mining and oil).
According to Lamb, Brazil remains a very “commodity” country in terms of exports. Changing this requires strengthening institutions and making them able to identify competitive advantages and prioritize from the point of view of science, technology and innovation. To build an innovation ecosystem in the 21st century, Lamb says it is essential to listen to the voices of academics from various fields, chart a new path that takes into account venture capital, and better places to invest, among other things. In 2022, the state’s planning includes an investment of R$112.3 million in innovation – the highest amount seen so far. He commented, “Science, technology and information can generate jobs, further develop cities, and improve the economy, so the government needs to realize this in order to keep progressing.”
“Back in the 1960s, we had a space race that cemented the United States as the country that led the planet for the next 60 years, because it demonstrated mastery of this new technology. Now, we are experiencing an economic and technological transformation where the question is: How, in this age of digitalization and artificial intelligence, Will we provide jobs that preserve human dignity?”, the academic asks. “Technology is not our destiny – it is a tool to shape our destiny.”
Advances in technology and their impacts on accessibility and health
Academic Henrik Malvar addressed the importance of artificial intelligence and new technologies to enhance access and improve the quality of health services. Distinguished engineer From Microsoft for nearly 25 years, Malvar has provided key products developed by (or in partnership with) the company. “The opportunity for digital transformation and economic development is primarily taken advantage of through artificial intelligence,” commented the researcher, who is a full member of the ABC and the Brazilian Academy of Engineering. He referred to Microsoft’s “AI for Good” program, which uses artificial intelligence in humanitarian actions such as healthcare and access to culture and accessibility. Among the main actions of the project, he mentioned applications to empower clinicians, making it possible to provide personalized care to the patient. Nuance EmpowerMD is the group’s big bet: clinical environmental intelligence, capable of revolutionizing medical consulting. The AI will listen to the conversation, produce a text – with keywords in bold – and synthesize a summary, suggesting possible remedies.
In the area of accessibility, he mentions automatic translations in the Microsoft Teams app, which provides translations in both the language being spoken and in languages other than the original voice—works for both those with hearing impairments and those who don’t know the language. In addition, he mentioned new technologies: adaptive control for people who do not have one hand and need to use a joystick (video game control), making it easier for these people to access modern video games; And the eye control tool in Windows 10-11, in partnership with Tobii, just looks at the screen.
Malvar’s main goal is to ensure equality and access to the benefits of AI for all, being a service that is low-cost and free of discrimination – as it has already happened, from algorithms that favor whites over blacks. AI, to function, needs to provide privacy, trust, inclusion, transparency, and accountability. You need to explain how AI came to a conclusion.”
Watch the full recording of the session:
Also available in English.