Sir Sadiq Khan previously warned AI could 'usher in a new ear of mass unemployment' for the capital

The vast majority of London jobs are at risk of being replaced by artificial intelligence (AI), a new report has warned.

The Organisation for Economic Co-operation and Development (OECEDt) has warned that three in four jobs in the capital are under threat from generative AI - more than any other city in the developed world.

That includes major US hubs such as New York and San Francisco, as well as European capital such as Paris and Berlin.

London has a high proportion of jobs in finance, professional services and creative industries - all fields which will be particularly affected by the rapid adoptions of AI.

Generative AI are already being used in software development and customer service, with agents used to perform complex tasks in a business.

The OECD classed three-quarters of London jobs as "highly-exposed" to AI - meaning more than half of their daily tasks could be automated away.

The Office for Budget Responsibility (OBR) has warned that 3.4 million people in Britain could lose their jobs due to AI in the next 10 years.

A further 30 per cent, or around 10.2 million people, could have their work complemented by AI within a decade, the OBR added.

The big four accountancy firms - KPMG, Deloitte, PwC and EY - have all said they are cutting junior employees because AI could complete work usually done by graduates.

Sir Sadiq Khan, the Mayor of London, warned earlier this year that AI could "usher in a new ear of mass unemployment".

"We need to wake up and make a choice: seize the potential of AI and use it as a superpower for positive transformation and creation, or surrender it and sit back and watch as it becomes a weapon of mass destruction of jobs," he added.

A spokesman for the Mayor said on Tuesday: "The Mayor believes that AI presents real opportunities – from driving economic growth to improving public services - but also brings with it new challenges, including the potential impact on London's labour market."

City Hall also created a London AI and Jobs Taskforce to examine how generative technologies had affected the capital.

The report follows a warning from Foreign Secretary Yvette Cooper that AI needs to be regulated to avoid a future earth-shattering display of power.

Comparing the emerging technology to nuclear power, she said international agreements on their regulation were made only after a display of "terrifying power".

"We cannot afford to wait for an AI equivalent of Hiroshima before we act," she said, in an essay written for Chatham House.

The Foreign Secretary said Britons were increasingly facing global instability, from rising energy and food prices to migration pressures and the risk of cyber attacks.

The Five Eyes intelligence alliance: the UK, US, Canada, Australia and New Zealand; warned AI-powered cyber attacks could be just months away.

Anthropic's newest version of AI model Mythos was not released to the public over fears it could be used to exploit software vulnerabilities.

On Monday it was revealed the US Cybersecurity and Infrastructure Security Agency (CISA) used Anthropic's model to audit Government software, sources familiar with the matter said.

Mythos is being used to detect potential bugs and vulnerabilities in the software, the source said.