Hundreds of public figures, including the co-founder of Apple and two ‘godfathers’ of AI, have banded together to call for a global pause on the development of superintelligent systems, warning that pushing on could risk human extinction.

Already signed by more than 900 leading lights from across tech, business, government, media, and faith (not to mention Prince Harry) the Statement on Superintelligence warns that the unchecked race to superintelligence among mega-AI firms is pushing humanity towards an uncertain future.

“Many leading AI companies have the stated goal of building superintelligence in the coming decade that can significantly outperform all humans on essentially all cognitive tasks,” says the statement.

“This has raised concerns, ranging from human economic obsolescence and disempowerment, losses of freedom, civil liberties, dignity, and control, to national security risks and even potential human extinction.”

Before creating a potenially world-ending bot, the statement calls for a prohibition on the development of superintelligence, not to be lifted before there is both broad scientific consensus it will be done safely and controllably, and the public is on-board with the risks.

Among those cautioning restraint are the so-called ‘godfathers of AI’, Yoshua Bengio and Geoffrey Hinton, whose work in AI and machine learning forms the bedrock of current technology.

In his statement in support, Bengio warned that frontier AI systems could surpass most humans across most cognitive tasks ‘within just a few years’ unlocking solutions for global problems, but first scientists and researchers need to design AI systems that are ‘incapable of harming people’.

Also signing on to stop a potential Skynet future are tech industry legends Steve Wozniak, co-founder of Apple, Peter Jensen, the inventor of the first ever mobile app, and Jaan Tallinn, co-founder of recently shuttered Skype.

The list of signatories goes on and on, with a host of names making for some strange bedfellows. Susan Rice, former US national security advisor under Obama, is in support, as is Steve Bannon, former chief strategist to Donald Trump.

The mix of policy makers from the left and right, of scientists and Catholic priests, actors, directors, royalty and rappers, signing on trying to slow down Silicon Valley’s superintelligence dream (slash nightmare), shows how far-reaching the anxiety around advanced AI has become.

But is there even anything to worry about? Despite major tech firms aiming to build superintelligent AI within the next ten years, and some forecasts saying it could be a reality by 2027, there still isn’t an agreed-upon definition of what AI superintelligence actually entails.

Recommended reading

• New Nonprofit PRISM Launched To Investigate AI Sentience
• OpenAI Co-Founder’s Safety-focused AI Startup Raises $1 Billion
• UK Shifts AI Focus to Prioritise Crime and Defence

Sam Altman himself has admitted that the definition of superintelligence and its precursor, AI general intelligence, has shifted, with what would once have been considered benchmark moments already being surpassed.

Elsewhere, critics argue that what we’re seeing now is nowhere close to superintelligence, with Brent Smolinski, IBM’s global head of technology, saying earlier this year, “It’s totally exaggerated. I don’t think we’re even in the right zip code for getting to superintelligence.”

Right now, AI still lacks the fundamental elements of human-like intelligence, relying instead on complex pattern analysis from unbelievably huge datasets to provide its answers, which, while undeniably lightning quick, are still not what scientists or philosophers would call ‘cognitive thought’.