What do the Pope, the Republic of Korea and the Labour manifesto have in common?

They're all talking about AI…..

Despite many politicians across the globe being in the midst of election campaigning, AI is still high on the political agenda. Even the Pope has a view – stating AI is ‘neither objective nor neutral’ and calling for political collaboration when addressing the G7 summit last week. 

Closer to home, the UK Government has been co-hosting the recent AI Seoul Summit 2024, and all major political parties have released their manifestos which include pledges around AI and other tech/digital issues. 

AI Safety Summit – take two

Last year’s AI Safety Summit at Bletchley Park was a big deal in the world of AI (see our blog), getting China, the US and a whole host of other nations and tech leaders together. 

On 21^st and 22^nd May the UK co-hosted the second iteration – the AI Seoul Summit 2024 - with the Republic of Korea. This latest Summit has led to a number of commitments, including:

• 27 countries committing to work together on thresholds for severe AI risks. 
• 16 AI companies from across the globe (including China and the US) signing “Frontier AI Safety Commitments”. 
• Leading AI developers agreeing to publish safety frameworks on how they will measure the risks of their frontier AI models (if they have not already done so) ahead of the AI Action Summit in France.  
• Countries pledging to boost international cooperation on the science of AI Safety and to support future reports on AI risk, following the publication ahead of Soul of the interim International Scientific Report of the Safety of Advanced AI. 

The next summit (the AI Action Summit in France) will be held early next year.

UK Election – manifesto updates  

On a national level, the UK political parties have also been making commitments around AI in their manifestos. 

Given its current lead in the polls, it is interesting to see that Labour’s manifesto mentions AI in a number of areas - although the detail is, unsurprisingly, fairly light. It has stated that it will ensure its industrial strategy supports the development of the AI sector and it will, for example, look to:

• Impose binding regulation on the ‘handful of companies developing the most powerful AI models’ (something which has already been suggested by the current government). 
• Ban the creation of sexually explicit deepfakes. 
• Establish a ‘Regulatory Innovation Office’, bringing together existing functions across government, helping regulators update regulation and co-ordinating issues that cross boundaries. The UK currently has the Digital Regulation Co-operation Forum but this only covers certain key regulators (the ICO, FCA, CMA and OFCOM). 

Its plans to reduce barriers to building digital infrastructure, particularly data centres, create a national data library to ‘help deliver data-driven public services’ and use AI in healthcare (to speed up diagnostic services) may also feed into AI growth. There has also been some discussion around whether Labour's plans to update employment laws could look into the impact of AI on workers, although this is not expressly mentioned in the manifesto.

Looking to the other parties, whose policies may be more or less influential depending on the extent to which current predictions of a Labour “super majority” prove incorrect:

• The Conservative's manifesto discusses securing the UK’s position as a world leader in innovation and building on its strong global position around AI security. It plans to double digital and AI expertise in the civil service, to take advantage of the latest technologies including AI and to transform public services (including investing £3.4 billion in new technology for the NHS). It will provide greater protection for creatives, while also emphasising the opportunities of AI, and will give the police power to use facial recognition technology to catch criminals. It will also continue investing in large-scale compute clusters, “assembling the raw processing power so we can take advantage of the potential of AI and support research into its safe and responsible use.” Its plans to urgently consult on introducing further parental controls over access to social media may also impact the AI sector. 
• The Liberal democrats discuss creating a clear, workable and well-resourced cross-sectoral regulatory framework for AI that: (i) promotes innovation while creating certainty for AI users, developers and investors; (ii) establishes transparency and accountability for AI systems in the public sector; and (iii) ensures the use of personal data and AI is unbiased, transparent and accurate, and respects the privacy of innocent people. Much of this appears to reflect current plans/rules, although the reference to resourcing addresses a concern with the current framework and regulator resource. It also plans to negotiate the UK’s participation in the Trade and Technology Council with the US and the EU, enabling the UK to play a leading role in global AI regulation and standardisation (for example they discuss working with international partners to agree common standards for AI risk and impact assessment, testing, monitoring and audit). National and local citizens assemblies will enable the public to engage on critical issues like AI and (like the other two parties) there are also plans for increased use of technology (including AI) in the NHS. 

Note: since this blog was published, the Labour party has won the UK's general election. It is expected that they will introduce an AI Bill in the King's speech on Wednesday 17th July.