The event appears to have succeeded, at least in starting the ball rolling in terms of regulation and oversight of AI technology.
The most advanced technology companies have agreed to allow governments to vet their AI tools before they are released on to the market. This will involve testing the safety of new models against dangers such as national security and societal harms.
Testing will be led by the US and the UK safety institutes - Prime Minister Rishi Sunak unveiled the UK’s AI safety institute prior to the summit with the remit to examine, evaluate and test new types of AI to better understand each new model. In fact, according to Sunak, many AI firms had already given the UK access to their models before their release.
There was also support for the setting up an international body, much like the Intergovernmental Panel on Climate Change, which will publish a ‘state of AI science report’ and will determine what AI can actually do.
Prior to the summit the UK also announced the Bletchley Declaration – a agreement signed by 28 countries including China, to work together on shared safety standards and testing.
However, the commitment to testing is voluntary and the UK government has baulked at bringing in legislation to rein in AI development and there was criticism of the event for allowing major tech players, such as Google DeepMind, to attend with warnings of a potential conflict of interest when it came to shaping AI regulatory processes.
The AI Safety Summit presented a unique opportunity for governments and experts to create a balance between effective regulation with safeguards and the needs of investors, businesses and consumers.
Earlier this year tech figures called for an immediate pause in the creation of “giant” AIs and warned that developers were, “locked in an out-of-control race to develop and deploy ever more powerful digital minds that no one – not even their creators – can understand, predict, or reliably control.”
This AI Summit has attempted to provide leadership and a clear plan to maximise AI's benefits while minimising its risks.
