US takes decisive action on AI safety

This move is set against the backdrop of a two-day summit hosted by the UK government at Bletchley Park, starting on 1 November, aimed at addressing concerns related to the swift progress of AI systems.

The executive order from President Biden mandates AI developers to disclose safety outcomes to the US government, positioning the US firmly in the epicentre of the global AI governance discourse. This has cast a shadow over the UK's ambition to assume this leadership role, especially given the timing of the summit it is holding.

Chatham House's senior research associate, Alex Krasodomski, expressed to the BBC that the executive order reinforces the perception of the US considering itself as the frontrunner in addressing such threats. White House Deputy Chief of Staff, Bruce Reed, further underscored this by stating that the US has introduced "the strongest set of actions any government in the world has ever taken on AI safety, security, and trust".

The executive order encompasses:

• Establishing new safety and security benchmarks for AI, inclusive of directives for AI firms to share safety test outcomes with the federal government
• Safeguarding consumer privacy through the creation of guidelines for agencies to assess AI-related privacy techniques
• Introducing measures to prevent AI algorithms from discriminating and laying down best practices concerning AI's role in the justice system
• Launching a programme to scrutinise potentially injurious AI-related healthcare practices and providing resources on the ethical use of AI tools in education
• Collaborating with global counterparts to standardise AI practices

Furthermore, the Biden administration is strengthening its AI talent pool, offering AI-specific job opportunities within the federal government via AI.gov.

Does this US action overshadow the UK's summit?

Krasodomski believes that while the order is of paramount importance, it might not align seamlessly with the UK's goals for the gathering. The UK's summit is indeed acknowledged in the US's executive order, but it falls under the ambit of 'advancing American leadership abroad', suggesting the US's dominant role in the AI space, especially with its corporate giants at the helm.

The UK government remains resolute in its aim to be a forerunner in mitigating risks associated with this potent technology. However, with the EU crafting an AI act, China having already implemented stringent AI regulations, and now the US's proclamation, the direction of the discussions at Bletchley Park might be influenced.

Furthermore, reports from the Reuters news agency indicate a forthcoming code of conduct for AI system developers by the Group of Seven (G7) industrial nations. This flurry of activity on the global AI governance stage raises pertinent questions regarding the substance and scope of the debates scheduled for the UK summit.

While the US's latest move signifies its intent to lead in the realm of AI governance, the UK, alongside other international players, will undoubtedly have a role in shaping the future trajectory of AI safety and governance. The outcomes from the Bletchley Park summit will be keenly observed in the days to come.