It has been revealed that AI firms have to take responsibility for all the harm they are causing before they can use AI in full force, according to the so-called godfathers of technology.
AI firms have to take responsibility for harm caused
A group of senior experts, including two “godfathers” of technology, have warned that AI companies must take responsibility for the harm caused by their products. They have also cautioned that the development of advanced AI systems without proper safety checks is utterly reckless.
The authors and academics have stated that powerful AI systems pose a threat to social stability. The warning comes ahead of an AI safety summit at Bletchley Park, where international politicians, tech companies, academics, and civil society figures will gather.
The co-author of the policy proposals made by 23 experts has said that it is important to understand how to make AI systems safe before pursuing more powerful AI systems.
“It’s time to get serious about advanced AI systems,” said Stuart Russell, professor of computer science at the University of California, Berkeley.
“These are not toys. Increasing their capabilities before we understand how to make them safe is utterly reckless.”
He added: “There are more regulations on sandwich shops than there are on AI companies.”
Document recommendations
The document recommends that governments adopt a set of policies that promote the safe and ethical use of AI systems.
These policies include allocating one-third of their AI research and development funding to such use, while companies should allocate one-third of their AI R&D resources as well. In addition, independent auditors should be granted access to AI laboratories and a licensing system should be established for the development of cutting-edge models.
Also, according to the document, AI companies must implement specific safety measures in case potentially harmful capabilities are found in their models. Lastly, tech companies should be held liable for any foreseeable and preventable harm caused by their AI systems.
Check out the original article posted by The Guardian in order to learn more details about all this.
