Experts have sounded alarm over the government's plan to create a portal using artificial intelligence (AI) to help people verify the information they receive, repel defamatory news and fight fraud or scams.
According to lecturer Benjamin Loh, there is a significant probability that this could result in further misinformation instead.
"Generative AI is infamous for providing unreliable information. Despite having vast amounts of training data, it is even far worse at handling contemporary fact-checking.
"It will likely produce more misinformation rather than correct it. This government has really bought the AI snake oil," said Loh, who is director of the Peace, Justice and Stronger Institutions Impact Lab at Taylor's University.
Therefore, he said it would be a better use of resources if the government advocated better digital media literacy training and awareness.
In making the announcement yesterday, Communications Minister Fahmi Fadzil said that the portal will expand the special segment “Biar Betul!” shown on the RTM channels.
He also said this would allow scam material, fake content and slanders to be exposed every day.
According to Fahmi, the new platform will strengthen the capabilities of existing platforms such as Sebenarnya.my under the Malaysian Communications and Multimedia Commission (MCMC), MyCheck.my managed by Bernama and “Biar Betul!” to make it easier for people to search and verify any information received.
Meanwhile, Loh also cited numerous failures of AI in the commercial realm, such as a case in 2022, when Air Canada's chatbot gave incorrect information to a traveller, resulting in a lawsuit that it lost.
He also pointed out that an AI engineer at Microsoft published a letter last month alleging that the company’s AI image generator lacks basic safeguards- against creating violent and sexualised images.
"Generative AI is mostly a smoke and mirrors show with much human labour involved in cleaning the outputs.
"It would be more useful if the government invested in fact-checking services and programmes instead," he said.
Kryss Network director Angela M Kuga Thas shares Loh's concerns, saying AI can create deep fakes and other forms of disinformation.
As such, she said it should not replace humans' critical thinking and assessment of the information they are fed through the information eco-system.
"AI has clear advantages to advance healthcare like making vehicles disability friendly and improving accessibility to healthcare.
"Don't misuse or abuse the deployment of AI, especially in relation to information and national security," she said.
Angela added that prejudices and discrimination are also part of human behaviour, which an AI can be trained to pattern, intensifying the biases and likely resulting in more injustices and human rights violations.
Kryss Network is a local organisation focusing on freedom of expression from a gender lens. - Mkini
No comments:
Post a Comment
Note: Only a member of this blog may post a comment.