However, much work is being done, often in the shadows, to try to harness their power in ways that strengthen democracy rather than destroy it.

“While this technology poses significant risks in terms of misinformation, it also offers significant campaigning opportunities, which we cannot ignore,” says Hannah O’Rourke, co-founder of Campaign Lab, a network of LEFT-focused tech volunteers.

“As with all technologies, what matters is how AI is implemented. Its impact will be felt in how activists use it.”

Campaign Lab in particular organizes training courses for Labor and Liberal Democrat activists on the use of ChatGPT (Chat Generative Pre-trained Transformer) to create the first version of election leaflets.

He reminds them, however, that they must carefully edit the final product, because large language models (LLMs) like ChatGPT have a worrying tendency to “blunder” or make things up.

The group is also experimenting with chatbots to help train direct sellers to have more engaging conversations at the door.

AI is already built into everyday programs, from Microsoft Outlook to Adobe Photoshop, says O’Rourke, so why not use it responsibly to free up time for more direct campaigns?

“AI-generated misinformation simply doesn’t work when politicians have good relationships with the voters and communities they represent, when they actually deliver on their promises.”

“You can’t pretend that the local park you pass every day has been cleaned, just as you can’t pretend to have a conversation on your front doorstep.”

Joe Reeve, an artificial intelligence expert who supports the Conservative Party, is a young activist convinced that new technologies can make things better.

He runs Future London, a community of “techno-optimists” who use AI to find answers to big questions like “Why can’t I buy a house?” and, above all, “Where is my robot butler?”

In 2020, Reeve founded Tory Techs, which is considered a right-wing replica of Campaign Lab.

The group runs programming sessions and uses AI to fine-tune Conservative campaign messaging, but Reeve says it is now focused “primarily on talking to MPs, in more private and secure spaces to help politicians understand what the “AI and how it can be a positive force.” .”

“Technology has the potential to make the world a much better place for many people, regardless of politics,” he tells BBC News.

The UK has already seen its first – and perhaps the world’s first – electoral candidate powered by AI.

Andrew Gray was roundly rejected by voters in the Selby and Ainsty by-election in July, winning just 99 votes.

It was a harrowing experience for the former lawyer, who ran as an “AI-powered” independent candidate, hoping to spark a political earthquake.

But voters in the rural North Yorkshire constituency were unwilling to accept such a “new” concept, he explains, and were often simply bewildered.

“When voters messaged me, they often thought I was a robot.”

Gray’s campaign relied on Polis, an AI-powered tool that allows groups with very different opinions to reach consensus through voting and debate.

Polis first appeared on the political radar when it was used by the Taiwanese government to resolve controversial issues, such as the legality of electric scooters.

Gray used it to develop a “people’s manifesto” for Selby and Ainsty and planned to use it to consult his constituents on how to vote in the House of Commons.

He firmly believes in Polis’s power to overcome political divisions and unite people, but believes he is perhaps better suited to resolving local disputes than tackling big national issues in Westminster, at least for the moment.

Gray’s charity, the Crowd Wisdom Project, has helped local councils, such as Wandsworth, south London, which held an air quality consultation last year, use Polis to formulate policy.

Critics will say that Polis and other AI-assisted decision-making programs marketed to local authorities are still susceptible to manipulation to produce the results desired by politicians, like the bogus consultation exercises that a previous generation of leaders loved.

But the sheer power of AI technology – and its ability to process vast amounts of data – promises to offer much deeper insights into what voters think and feel.

Joe Twyman, co-founder of polling firm Deltapoll, says sophisticated “regression” polls – of the kind YouGov recently produced to predict a landslide Labor Party victory in the general election – were only possible because of AI.

According to the British Polling Council, the average regression survey has between 6,000 and 10,000 respondents.

Deltapoll works with startup Bombe, which uses artificial intelligence to analyze survey data collected from smaller samples.

Twyman is quick to reject the idea that LLMs can be used to simulate voters.

In 2022, American researchers claimed that the results were indistinguishable from those of real humans taking a survey.

But Twyman believes that people are more complex and unpredictable than machines trained on existing data and that “synthetic surveys” would have “no value.”

He admits, however, that it is impossible to know exactly which of the growing number of AI tools offered to political campaigns will be adopted.

“Some of them will be incredibly powerful, some of them will be completely absurd and promoted by snake oil salesmen exploiting the next big thing, and there will be things we will never hear about that could potentially make a big difference,” Twyman says. .

Most tech optimists worry about the ability of AI-powered disinformation to wreak havoc on elections, especially since by 2024, an estimated 49% of the world’s population will turn out to vote.

The benefits to democracy of this powerful and rapidly evolving technology may seem marginal compared to the harm it could cause.

But no one can predict how this will play out: the only certainty is that AI is here to stay and politics will never be the same.