The safeguards included in the UK government’s AI white paper are “inadequate”, according to the Equality and Human Rights Commission (EHRC).
The non-departmental government body said that current proposals to regulate AI “fall short of what’s needed to tackle the risks to human rights”.
The remarks will come as a blow to Prime Minister Rishi Sunak, who has said he wants the UK to be the “geographical home of global AI safety“.
While the EHRC acknowledged the benefits AI can bring, it said there must be a greater focus on how it will impact equality.
The equality watchdog also recommended increasing funding for regulators, including itself, to manage the rapidly advancing technology.
“People want the benefits of new technology but also need safety nets to protect them from the risks posed by unchecked AI advancement,” said Baroness Kishwer Falkner, chairwoman of the EHRC.
“If any new technology is to bring innovation while keeping us safe, it needs careful oversight. This includes oversight to ensure that AI does not worsen existing biases in society or lead to new discrimination.
“To rise to this challenge, we need to boost our capability and scale up our operation as a regulator of equality and human rights. We cannot do that without government funding.”
The EHRC, however, acknowledged the white paper as a step in the right direction, and praised the government on its ambition to develop a robust regulatory framework for AI.
The government has faced a mixed response to its AI regulatory white paper, with some praising the sector-specific measures, while others have questioned the gaps in the published material.
Labour MP Darren Jones, chair of the Business and Trade Committee, called on the government to explain what action it would take to ensure safety in AI.
The government responded by saying it would continue to develop safety measures and work with international players to ensure protections are put in place.