Last spring, in the wood-panelled rooms of Manchester Art Gallery, a group of artists, philosophers and curious members of the public gathered for a Philosophy Café session and asked a question that, at the time, felt almost academic: should we fear or welcome artificial intelligence? They stood in Gallery 11 beside a small William de Morgan tile depicting a dodo – an emblem of extinction – and wondered whether humanity was witnessing its own creative obsolescence. Eighteen months later, the question is no longer philosophical. It is legislative, economic and existential. The creative industries – worth £146 billion annually to the UK economy – are locked in a confrontation with a technology that promises to elevate them and threatens to hollow them out simultaneously. And the answer, increasingly, is neither fear nor welcome but something far more combative: organised resistance.
The Fear That Refused to Subside
When Katerina Karp, a visual artist and researcher, reported on that Manchester Art Gallery session for Creativity’s UK in April 2025, the dominant emotion she identified was fear of the unknown. AI, she observed, is unlike the steam engine or penicillin – it can be applied virtually everywhere, and that boundlessness is its most unsettling quality. The participants agreed that panic about a machine uprising was misplaced; at the start of every technological process stands a human being endowed with critical thinking. But they also conceded that AI had already made it easier to spread fake news, blur the boundaries between authentic and generated content, and raise questions about fairness in access to its power.
Those fears have not dissipated. They have sharpened. In March 2025, the Design and Artists Copyright Society published research indicating that a third of UK artists are already using AI in their practice, while simultaneously voicing significant concern that unregulated AI would negatively impact their careers, future opportunities and copyright. The speed of adoption has outpaced the development of ethical frameworks, leaving creators exposed. As Lucy Brooke, a Hull-based curator and artist, told the BBC in March 2026, the growing anxiety centres on “copyright, plagiarism and exploitation” – concerns that prompted her to organise free community sessions to help artists understand AI’s environmental, political and social impacts. The overwhelming view from creatives, she said, was that the answer lay in collective action, not individual retreat.
Artists Are Not Waiting for Permission
The popular image of the artist paralysed by technological change is contradicted by the evidence. According to the UK government’s June 2026 AI Adoption Plan for the Creative Industries, 51 per cent of creative businesses already use AI – compared with 33 per cent across all UK businesses. The technology is most prevalent in IT and software (60 per cent), design (53 per cent) and film and television (44 per cent). What is striking is that adoption is happening from the ground up: smaller firms and freelancers are experimenting with AI tools for ideation, production workflows and audience engagement, often without formal governance structures in place.
A 2026 study by Spark AI found that in more than half of UK creative agencies, AI usage is happening with no formal policy – no whitelisted tools, no data-handling training, no institutional guardrails. Staff are innovating alone, carrying their agency’s legal and intellectual property exposure. The image of the solitary artist wrestling with a text prompt is, in reality, a picture of an entire workforce improvising in real time. The question is whether this ad hoc revolution will produce a renaissance or a crisis of accountability.
Westminster Enters the Fray
The political terrain has shifted dramatically since those Manchester participants first asked whether humans could influence AI’s development. In March 2025, the UK government launched a public consultation on copyright and AI, proposing to allow technology companies to train their models on copyrighted material unless the rights holder explicitly opted out. The proposal ignited a firestorm. Sir Elton John, among more than 400 musicians, writers and artists who signed a letter to Prime Minister Sir Keir Starmer, called the plans “criminal” and accused the government of “committing theft, thievery on a high scale”. “The danger is for young artists,” he told the BBC in May 2025. “They haven’t got the resources to keep checking or fight big tech.”
The House of Lords intervened with unusual force. In March 2026, the Communications and Digital Committee published its report, AI, Copyright and the Creative Industries, which warned that the UK faced a choice between two futures: one in which the nation becomes a world-leading home for responsible, licensing-based AI development, and another in which it drifts towards tacit acceptance of large-scale, unlicensed use of creative content. Baroness Keeley, the committee’s chair, was unequivocal: “We are calling on the government to embrace the opportunities this presents, and to demonstrate its commitment to the UK’s gold-standard copyright regime and our outstanding creative industries.”
The government’s position collapsed shortly thereafter. In a report published on 18 March 2026, the Department for Science, Innovation and Technology conceded that the broad copyright exception with opt-out was “no longer the government’s preferred way forward”. Science minister Liz Kendall acknowledged the creative industries’ overwhelming rejection of the proposal. Tom Kiehl, chief executive of UK Music, welcomed the reversal but was blunt about the damage already done: “The 220,000 people in our sector which generates £8 billion for the UK economy should be entitled to work and earn a living without the constant fear that the fruits of their labour could effectively be taken by AI firms without payment or permission.”
The Environmental Ledger
What the parliamentary debates often obscure is AI’s physical cost. Training a single large language model can consume as much electricity as a small town uses in a year. At the Manchester Philosophy Café, the group watched a short film about AI and climate change and immediately raised the question of resource consumption and fair distribution. Could AI democratise access to climate knowledge and solutions? Or were corporations exaggerating their role, claiming that only through them could the world tackle the climate crisis?
The Association of Illustrators’ Brave New World report, published in early 2026, was blunt: generative AI technology, “unleashed without regulation, safeguards or guardrails, is rapidly disrupting the UK’s £124.6 billion creative sector.” The report documented that 99 per cent of creators said their work had been scraped without consent, and that the environmental footprint of AI training directly contradicted the UK’s net-zero commitments. The creative sector, which supports more than 2.4 million jobs, faces what the report describes as a process of “digital de-industrialisation”, as economic value shifts from human creators into the hands of multinational technology companies.
When Machines Make Art, What Counts?
The question that most animated the Manchester gallery participants – can we assess the value of a piece of art in the moment? – has acquired a new urgency since Ai-Da Robot, a humanoid artist built by a team led by Oxford gallerist Aidan Meller, sold a portrait of Alan Turing at Sotheby’s in November 2024 for $1,084,800. The sale price smashed its estimate of $120,000 to $180,000, drawing 27 bids and prompting Sotheby’s to declare it a moment “in the history of modern and contemporary art”. The painting depicted Turing as “the god of artificial intelligence” – a recursive tribute from the machine to its conceptual father.
Ai-Da told the House of Lords Communications and Digital Committee in 2022: “I do not have subjective experiences; I am dependent on computer programmes… although I’m not alive, I can still create art.” The statement encapsulates the paradox. If a robot can produce work that the market values at seven figures, what happens to the human artist whose training, emotional life and cultural memory informed their practice? The Manchester group argued that AI’s hallmark is its ability to learn and adapt – a human trait we may have found a rival for. But human learning is tied to emotion and belief; AI learns through logic, devoid of feeling. The distinction may prove to be the last redoubt of human creative authority.
Manchester’s Unlikely Centre of Gravity
It is no accident that the question of AI and art keeps returning to Manchester. The city where the world’s first stored-programme computer was built is now host to some of the most rigorous interrogation of AI’s cultural implications in the country. The CreaTech Network Series, a three-part programme running between February and June 2026, is bringing together researchers, creatives and technologists to examine AI in entertainment, cultural heritage and open-source tools. The School of Digital Arts at Manchester Metropolitan University is staging Storytellers + Machines 2026: The End of AI, a conference that asks whether the creative disruption of generative AI is a passing phase or a permanent transformation.
And in the Philosopher’s Café at Manchester Art Gallery, the conversations continue. The monthly sessions – free, unstructured, open to anyone who wishes to listen and respect other people’s points of view – remain a space where the abstract becomes personal. As Karp observed in her original account, everything revolves around studying human behaviour. Perhaps we are approaching a moment of truly decoding ourselves – and we might not like what we find. The creative industries are not waiting for that moment. They are organising, legislating, litigating and making art in spite of it. The dodo tile in Gallery 11 is no longer a metaphor for extinction. It is a reminder that survival belongs to those who adapt – but on their own terms.
