The UK and the US, along with 16 other countries , have released new guidelines that companies must follow to develop safe artificial intelligence (AI) systems.
“The approach prioritizes ownership of security for customers, embraces full transparency and accountability , and creates organizational structures where secure design is a top priority,” said the U.S. Cybersecurity and Infrastructure Security Agency (CISA).
The National Cyber Security Centre (NCSC) says the guidelines aim to levels cybersecurity of AI to ensure the technology is designed, developed and implemented in a secure manner.
See also: Ipsos Consumer Tracker survey on artificial intelligence
The new guidelines also build on the US government’s efforts to manage the risks posed by artificial intelligence (AI). It has been decided that it must ensure that new tools are adequately tested before public release and that there are safeguards in place to address potential harm to society . For example, it must ensure that there is no bias and discrimination and that any be addressed concerns privacy must . Finally, it must ensure that consumers are able to identify material generated by AI.
Additionally, the new rules require companies to commit to allowing and facilitating third parties to discover and report vulnerabilities in their AI systems . Essentially, companies will have to have bug bounty programs so that researchers can find security holes. This will allow for immediate fixes, while researchers will receive a monetary reward for their work.
See also: Artificial intelligence (AI) and surveillance technology
The latest guidance “helps developers ensure that cybersecurity is both a fundamental prerequisite for security and an integral part of the development process from the outset and throughout, an approach known as secure by design,” the NCSC said.
This includes secure design, secure development, secure implementation, and secure operation and maintenance, covering all important areas in the AI systems development lifecycle. Companies should model threats to their systems as well as protect their supply chains and infrastructure.
Also, the security services of the countries participating in this initiative say that it is important to combat attacks targeting artificial intelligence and machine learning (ML) systems with the aim of carrying out unauthorized actions and stealing sensitive information.
New guidelines for developing secure Artificial Intelligence (AI) systems from the US, UK and global partners will revolutionize AI security
See also: Many IT professionals are negative about the use of artificial intelligence (AI)
They will allow developers and companies to develop AI systems that are more resistant to attacks and protect user privacy and security.
Furthermore, the guidelines will promote transparency and accountability in the development of Artificial Intelligence, ensuring that the algorithms and decisions of AI systems are fair and trustworthy.
Source: thehackernews.com
