The US and the UK have launched a cooperative effort focused on artificial intelligence (AI) safety, in response to concerns about the impact of advanced AI technologies.
Commerce Secretary Gina Raimondo and British Technology Secretary Michelle Donelan signed an agreement in Washington to collaborate on testing advanced AI models.
This agreement, part of commitments from a previous AI Safety Summit, aims to address national security risks and broader societal concerns related to AI.
Both countries are initiating government-run AI safety institutes, with the UK planning to scrutinize new AI types and the US evaluating risks from advanced AI models, in collaboration with 200 entities.
The partnership includes plans for joint testing exercises and potential staff exchanges between the institutes, with an aim to extend similar cooperations internationally to enhance AI safety globally.
The collaboration is seen as a vital step in harnessing AI's benefits while mitigating risks, including job displacement, election interference, and potential human obsolescence.
In light of the imminent release of more sophisticated AI models, Raimondo and Donelan emphasize the urgency of addressing AI risks comprehensively. Raimondo will discuss
AI concerns further at an upcoming US-EU meeting in Belgium, with the Biden administration also expanding its AI team to bolster efforts in this area.
The US and UK commit to sharing crucial data on AI capabilities and risks, along with conducting technical research on AI safety. Recent government actions, including executive orders and funding for AI research hubs and regulatory training, reflect the seriousness of AI threats, including its potential use in bioterrorism and nuclear simulations.