Before the upcoming event in Seoul, Rishi Sunak announced that the initial 16 companies had agreed to adhere to the voluntary artificial intelligence safety standards introduced at the Bletchley Park summit.
Critics claim that the standard lacks effectiveness, as signatories only commit to information sharing, investing in cybersecurity, and prioritizing social risk research voluntarily. “That’s our only promise,” Sunak stated.
Sunak mentioned, “These efforts are aimed at ensuring that the top AI companies globally offer transparency and accountability in their plans for developing secure AI. This will establish a precedent for a global AI safety standard that maximizes the benefits of this innovative technology.”
The 16 companies include Zhipu.ai from China and the Technology Innovation Institute from the United Arab Emirates. The government sees the participation of signatories from countries hesitant to enforce safety regulations as an advantage of a lighter approach.
UK Technology Secretary Michelle Donovan praised the Seoul event for advancing the work started at Bletchley, creating the “Bletchley effect” thereafter. She highlighted the impact on the global AI safety agenda, with countries planning to establish their AI safety laboratories.
Regarding the voluntary nature of the code, Fran Bennett, interim director of the Ada Lovelace Institute, cautioned that without regulations, AI companies might disregard it entirely.
After newsletter promotion
Later today, Sunak will co-chair a virtual meeting with South Korean President Yoon Seok-yeol and global leaders on “innovation and inclusion” in the AI field.
Source: www.theguardian.com
