Elon Musk and other tech experts call for a 'pause' on advanced AI systems

More than 1,000 executives and researchers in the tech industry have called for a six-month pause in the development of advanced artificial intelligence systems such as OpenAI's GPT in order to stop what they believe is a "dangerous" arms race.

‍

Within hours of the Future of Life Institute, a non-profit advocacy group, releasing an open letter on Wednesday, more than 1,100 individuals from academia and the IT industry signed it.

‍

“In the recent months, there have been numerous reports that AI labs have been locked in an out-of-control race to come up with ever more powerful digital minds that no one - not even their creators - has been able to predict or reliably control,” said the letter.

‍

The letter was also signed by Stuart Russell and Yoshua Bengio, the two most renowned AI professors and the co-founders of Apple, Pinterest, Skype and the AI start-ups Stability AI and Character.ai. The Future of Life Institute, which published the letter, is headed by Max Tegmark, a professor at the Massachusetts Institute of Technology and an expert in artificial intelligence. Musk is one of the organization's major funders.

‍

"We demand that all AI labs immediately halt the development of AI systems more potent than GPT-4 for at least six months. This halt should involve all relevant players and be open to verification.

‍

This halt should involve all significant players and be made public and verifiable. Governments should intervene and impose a moratorium if such a pause cannot be immediately implemented," the group added.

‍

The letter comes in the wake of a wave of ground-breaking AI product debuts over the last five months, including Microsoft-backed OpenAI's ChatGPT in November and this month's release of GPT-4, the complex model that powers the chatbot.

‍

Millions of regular consumers now have access to AI thanks to actions taken by businesses like Google, Microsoft, and Adobe that have included new kinds of AI functionality in their search engines and productivity applications.

‍

Some AI researchers and tech ethicists are concerned about the possible effects of this rapid rate of development and public deployment on employment, public discourse, and — eventually — humanity's capacity to keep up.

‍

The letter proposed the development of common safety standards that are subject to audits by impartial experts in order to "ensure that systems adhering to them are safe beyond a reasonable doubt."

‍

"Human-competitive AI systems might seriously endanger society and humanity," it stated.

‍

One of the leading names on the growing list of signatories to the letter is noted AI scientist Bengio, professor at the University of Montreal, and Berkeley professor Russell. 

‍

As a former co-founder of OpenAI, Musk left the organization in 2018 and since then has become a critic of OpenAI, which he has signed the letter as well. The list also includes Apple co-founder Steve Wozniak, author Yuval Noah Harari, and former US presidential candidate Andrew Yang.

‍

Several engineers and researchers from Microsoft, Google, Amazon, Meta,, and Alphabet-owned DeepMind are also part of the project. The letter was signed by only 1,000 people, none of whom identified themselves as employees of OpenAI when signing it.

‍

As governments around the world race to formulate policy responses to the rapidly evolving field of artificial intelligence, even as some of the world's largest technology companies are reducing the amount of resources dedicated to AI ethics teams, the intervention comes at a critical time.

‍

In a white paper published on Wednesday, the UK will ask existing regulators to develop a consistent approach to the use of artificial intelligence in their respective industries, including ensuring that AI systems are transparent and fair in their use. To date, the government has not indicated that it would provide new powers to regulators or fresh funding at this stage of the process. 

‍

A set of rules governing the use of artificial intelligence in Europe is also being prepared by the EU, and companies that break these rules will be punishable with fines of up to €30mn or six per cent of their global annual turnover, whichever is greater.