Approach the topic of AI with 'considerable cynicism', expert says

The Oireachtas Committee on AI will hear that the more pertinent impacts of the technology are those currently seen on social media, both in terms of creating engagement via content and the algorithms that dictate what users see. File picture

The public should treat the concept of any news about artificial intelligence with “considerable cynicism”, an Irish expert on the technology has said.

Professor Barry O’Sullivan of the school of computer science and IT at UCC will tell the Oireachtas Committee on AI that, despite global consternation at the prospect of whole livelihoods and sectors being rendered redundant by AI, “major obstacles still exist to building systems that really compete with the capabilities of humans".

He will tell the committee, which has been considering the societal impacts of AI, that while there is a lot of discourse around concerns, such as that AI will create massive job displacement, the more pertinent impacts of the technology are those currently seen on social media, both in terms of creating engagement via content and the algorithms that dictate what users see.

“I would encourage policy-makers, investors, and the general public, to approach the topic with considerable cynicism,” Prof. O’Sullivan will tell the committee.

“We are frequently presented with extreme risks such as massive job displacement, a fundamental transformation of society, and even the possibility of an existential threat to humanity,” he is expected to say.

“However, inadequate focus is dedicated to clear and present dangers such as the impact of racial and gender bias, the impacts on mental health, and the broader societal impacts of AI-enabled social media platforms both in terms of content creation and targeting content delivery.”

The professor, an expert and consultant on the military application and utilisation of AI, will tell the committee that while there is a great deal of consensus internationally regarding the need for limitations on fully-autonomous weapons systems such as those with nuclear capability, there is “less agreement” regarding the semi-autonomous use of such systems.

He will emphasise the need to find “common ground” between nations on limitations to AI within military systems before such systems “become so commonly used as to make future limitation impractical”.