x

Abu Dhabi, UAEThursday 13 December 2018

AI must not be used for 'terrifying' new weapons systems, Abu Dhabi event hears

Tech companies and governments need to show 'ethical leadership' over artificial intelligence

Scientists and campaigners are increasingly concerned that cutting edge artificial intelligence could be used to make deadlier weapons, of which the fictional killer Terminator must be the most infamous. EPA
Scientists and campaigners are increasingly concerned that cutting edge artificial intelligence could be used to make deadlier weapons, of which the fictional killer Terminator must be the most infamous. EPA

Technology companies and governments should strike new partnerships to ensure artificial intelligence is not harnessed for “terrifying” purposes including new weapons systems, experts have said.

In a panel discussion the audience was warned about the potential for their minds to be controlled and false memories implanted by advanced technologies and that these inventions could potentially be weaponised.

Sophisticated face-swapping and voice mimicking technology could also be used for sinister ends, for example by impersonating politicians and convincing the public they had make speeches they had not, it was warned.

Politicians were urged to open a deeper dialogue with the companies developing the new systems, to ensure that the many potential benefits of innovations such as artificial intelligence and robotics are captured and the risks minimised.

Helene Holm-Pedersen, an adviser for the European External Action Service, the EU's foreign and defence ministry, said there was an appetite among tech companies for governments to step in to provide ethical leadership. Some workers developing new technologies, she said, were increasingly becoming concerned about how their inventions could be misused.

“Is there a common interest in trying to contain some of the worst abuses? I think there is,” she said. “We are entering a world where not only are tech actors becoming foreign policy players but they are also looking for us to come back into the world of technology and help provide ethical leadership, public leadership.”

She said a conversation should be held between tech companies, government and civil society to solve problems around both development of positive new innovations, as well as those which have “terrifying implications if left unchecked”.

“In a sense we need to have a new conversation between those who are leading the charge, developing the new technologies, and those of us who represent governments in order to get the diagnoses right, about the dangers and opportunities,” she said.

“I think we need to repersonalise technology. Many big tech companies, which are putting on the market the technologies, there are people behind those who are often quite concerned about some of the dual uses of the technologies they are developing too.”

A ‘robot’ activist for the Campaign to Stop Killer Robots in London's Parliament Square. Getty Images
A ‘robot’ activist for the Campaign to Stop Killer Robots in London's Parliament Square. Getty Images

The EU recently launched a new global tech panel, aimed at bringing industry, civil society, acadmics and government together, as well as fostering ties with diplomats. Its aim is to start a conversation about not just tackling the threats of the digital age, but also unleashing its potential to solve global problems in partnership.

Jean-Marc Rickli, head of global risk and resilience at the Geneva Centre for Security Policy, said the rise of artificial intelligence had particularly important implications for society. “What I’m concerned about is the rate at which this technology evolves, which is exponential,” he told Diplocon delegates.

New developments, such as the ability to change faces in moving images and voice mimicking technology, could have a “tremendous impact” on diplomacy, he said, as they could be used for sinister purposes.

“On the screen you could have President Obama talking, but it would be my words. Most people will tell you I only believe what I see. Nowadays the problem is that what you see can completely be forged.

“This is just an introduction, I’m not talking yet about autonomous weapons systems that goes beyond your imagination or the fact that in the very near future, we will be able to pair your brain with an algorithm, and new categories of weapon, which involve controlling your mind from a distance, will be possible.”

_________________

Diplocon reports:

War games simulate India and Pakistan nuclear crisis at UAE summit

Minister fears voices of intolerance becoming louder across the globe

Nations must break up the diplomatic boys' club, UAE summit hears

Gargash: in the 15-minute news cycle, we can't afford to ignore criticism

Middle East states 'need to speak to their enemies'

_________________