Top researchers suggest that governments and companies should increase their investment in ensuring the safety of artificial intelligence.
Top artificial intelligence researchers have recommended that both companies and governments dedicate at least 33% of their AI research and development budget towards guaranteeing the safety and ethical implementation of these systems. This suggestion was outlined in a paper published on Tuesday.
A document released one week prior to the international AI Safety Summit in London outlines actions that governments and businesses should implement to address the risks associated with AI.
According to a paper authored by three recipients of the Turing Award, a Nobel Prize winner, and numerous esteemed AI researchers, it is necessary for governments to require companies to be accountable for any damages caused by their frontier AI systems that could have been predicted and avoided.
At present, there are no widespread rules specifically addressing the safety of AI. The initial legislation proposed by the European Union is still pending as policymakers have not reached a consensus on various matters.
Yoshua Bengio, known as one of the “godfathers” of AI, stated that the current advanced AI models are too potent and impactful to be left to develop without democratic supervision.
According to him, we must act quickly on investing in AI safety since AI is advancing at a rapid pace and precautionary measures are lagging behind.
Some of the authors are Geoffrey Hinton, Andrew Yao, Daniel Kahneman, Dawn Song, and Yuval Noah Harari.
Ever since OpenAI introduced generative AI models, leading scholars and notable CEOs like Elon Musk have expressed concerns about the potential dangers of AI. Some have even proposed a six-month hiatus in the development of highly advanced AI systems.
Several businesses have responded by stating that they will incur significant expenses to comply and face unequal levels of liability.
British computer scientist Stuart Russell stated that companies often argue that regulations are too difficult to comply with and that they stifle innovation, which he believes is absurd.
Sandwich shops are subject to more regulations compared to AI companies.