AI industry needs more people of colour, says at World Government Summit

Programmers with more diverse backgrounds are needed to fix algorithmic biases

Beta V.1.0 - Powered by automated translation

Generative artificial intelligence technologies like ChatGPT will be “fruitful”, but more work is needed to get rid of algorithm biases, pop star has said.

The Black Eyed Peas singer and owner of tech start-up FYI, said that the tech field needs more people of colour that can do programming to reduce algorithm biases.

He was speaking on Monday at the first day of the World Government Summit in Dubai, at a panel session titled Accelerating Tech: The New Frontier for Policy-making.

“It’s exciting times especially when it comes to generative AI,” said

“That being said a lot of jobs will be rendered obsolete, but new jobs, careers and industries will sprout.

“It’s going to get more fruitful when other companies release their generative AI. GPT is awesome and they’re the first step in this field.

“It’s going to be a very fruitful playing field for creatives to choose a selection of generative AIs.” at the 'Accelerating Tech: The New Frontier for Policy-making' session at the World Government Summit in Dubai. Pawan Singh / The National

ChatGPT is a chat bot developed by OpenAI that interacts in a conversational way.

So far, it has passed exams for law and business schools, and has almost passed the US medical licensing exam.

Google has launched its own version of the chat bot, called Bard, with other global companies also following suit.

“One [challenge] is algorithm biases and it’s the reason why we do the work philanthropically at our foundation.

“It’s to inspire young kids that come from inner city, people of colour, to write algorithms and train in data because the only people that are going to fix algorithmic biases are people who look like me.”

A 2022 report by the European Union Agency for Fundamental Rights showed that Dutch tax authorities had used algorithms that mistakenly labelled about 26,000 parents as having committed fraud in their childcare benefit applications.

Many of these parents had an immigration background and were required to pay back large sums.

The report said that the data protection authority had concluded that the processing of data by the AI system in use was discriminatory.

Doron Avni, Google's vice president of government affairs and public policy for emerging markets, was also speaking at the panel.

He said that before Google could officially release Bard, the company needs to make sure they “get the quality right”.

“It' important to get it right. That's why we announced that we are sending it out to trusted testers,” he said.

“Before we launch it out to the public — and this will happen in a few weeks — we want to get it right in the sense of quality, responsibility and being groundedness in real world data.”

He said that governments and private companies should be working together for innovation to happen, but there also need to be clear guidelines.

Mr Avni also spoke about how Google was trying to use AI in ways that could benefit humanity.

“We're doing some collaboration with governments and using AI to make the world a better place. We have flood forecasting tools that use complex AI models,” he said.

“We're also sending texts on floods to people on the ground who could be affected.”

Updated: February 15, 2023, 12:15 PM