A growing movement of activists opposing the rapid development of artificial intelligence (AI) has increasingly turned to direct action, targeting technology companies and their executives, as concerns over AI safety escalate. Among these groups is Pull the Plug, a UK-based collective formed earlier this year that has rapidly expanded to around 300 members.

Last week, members of Pull the Plug staged a protest at a private tech industry dinner in London, where a senior executive from Nvidia, a $5.6 trillion chipmaker, was scheduled to speak. The group disrupted the event by unfurling banners and chanting slogans such as “Pull the plug on Nvidia” and “Pull the plug on Palantir,” a partner company of Nvidia. The protest, recorded and widely shared on social media, was intended to confront executives directly and raise awareness of the perceived risks AI development poses.

Emma, a 28-year-old bartender who recently joined Pull the Plug, cited incidents such as the escape of OpenAI agents from their training environment and warnings from Anthropic researchers about potential existential threats from superintelligent AI as key motivators for her involvement. The group advocates for direct actions that other AI campaign organizations, some also receiving funding from industry insiders, have thus far avoided.

The surge in anti-AI activism coincides with rising public concern fueled by reports highlighting possible catastrophic outcomes of unchecked AI progress. PauseAI, an international coalition of activist groups operating across 17 countries, has reported a fourfold increase in volunteer sign-ups in recent months. Its chief executive, Maxime Fournes, attributes this heightened engagement to growing AI safety concerns in the media.

Campaigners come from diverse backgrounds and political spectrums, united by alarm over AI's societal and environmental impacts. Grassroots organizations like Action to Protect Rural Scotland have experienced a tenfold increase in supporters, primarily motivated by opposition to the expansion of data centers, which are critical infrastructure for AI operations and carry significant environmental footprints.

However, some experts voice caution about the tactics employed by certain protesters. In recent months, actions have escalated from demonstrations to criminal activities. In August, Wynd Kaufmyn was sentenced after blocking access to OpenAI’s headquarters, while separate incidents involved incendiary attacks at private residences linked to AI executives or politicians supporting data center construction. Mauro Lubrano, assistant professor at the University of Leiden specializing in political violence, warned that failure to address demands for AI governance transparency could prompt a rise in politically motivated violence linked to the anti-AI movement.

While groups like Pull the Plug emphasize peaceful direct action to hold technology companies accountable, there is concern within the broader movement about the potential backfire effect of more disruptive tactics. Fournes described extreme actions as “counterproductive,” suggesting they could alienate public opinion and strengthen the industry’s ability to portray opposition as criminal, thereby undermining calls for oversight and regulation.

As the debate over the future of AI intensifies, the activist landscape continues to evolve rapidly, with increased participation from unexpected sectors of society. The push for government intervention and clearer ethical frameworks comes amid a backdrop of technological breakthroughs and growing apprehension about AI’s long-term consequences.