Do they think AI will kill us, or do they desperately need public support?

Last week, Anthropic engineers, their leaders, and Elon Musk's tweets now agree that there is a greater than 10% chance that AI Large Language Models (LLMs) have the potential to kill humans - yet those LLMs are still live, making money. All the companies have done so far is publish 'safety' planning documents on X, paying lip service to these concerns.

But why is this such a sudden concern? It's no secret that both Anthropic (Claude) and OpenAI (ChatGPT) are in the late stages of preparing for their IPOs, falling slightly behind Grok's IPO (as part of SpaceX). We could read these comments as an effort to look responsible to investors at the last moment - proof that this technology is not only massive and disruptive, but that we should continue to trust these men to lead it. Look, they can write guidelines that they may or may not stick to.

Aside from the IPOs, the timing of this concern feels interesting to me. The public at large is beginning to despise AI and its resource usage, pushing back on data centres.

This summer, the planet suffered yet another season of extreme weather events, including drought, heatwaves and wildfires dominating the news across Europe. This extreme heat and drought drew a lot of attention to data centres, the physical infrastructure behind 'AI tools'.

In Ireland, data centres consume 23% of the country's electricity, raising questions about their water use. Water usage is harder to measure for data centres because they have different water requirements. However, they typically need fresh (or drinking) water to help cool the ginormous rows of servers. Some data centres get their water from state/private provider(s), some do this infrequently using closed-loop systems for cooling, and others store groundwater in lakes or wells on site.

This question became urgent in July 2026 in Ireland, when Uisce Éireann, Ireland's state-owned water provider, warned residents in Leinster that they would have reduced water and, in some cases, no water overnight for a period due to low reservoir levels. Public anger grew; was this a two-fold issue? The lack of rainfall (July had only 17% of Ireland's average rainfall), but could it also be due to increased data centre cooling?

While some argued this was insignificant, Uisce Éireann quickly claimed it had no idea how much water data centres were consuming. Shortly after, Uisce Éireann announced a country-wide hose-pipe ban (encouraging residents to rat out neighbours and businesses who broke it). This quickly turned people's attention back to data centres: 'Can we rat on them for using our water?' 'Can we rat on Jimmy's community group using AI to make a poster?' The frustration that data centres are draining Ireland of its resources in a time of climate crisis began to gain more and more attention.

Ireland is not the only place seeing this pushback or asking difficult questions about the rapid growth of AI infrastructure, which requires larger data centres; New York, Tennessee, and many cities across Europe are starting to rise up. That said, Ireland's electricity use by data centres is the highest per capita.

Ireland is, however, 'lucky' that these centres are only taking electricity from a fossil fuel grid, unlike the horrors residents in Tennessee are being subjected to by Grok's off-grid centres. As of writing, there is now a risk that Irish data centres could increase their off-grid capabilities as the Commission for Regulation of Utilities proposes cheaper bills for centres that agree that they aren't guaranteed 24/7 undisturbed access to the grid.

The depleting resources, the risk of grid failure, the risk of water outages, and the ongoing destruction to communities, including reduced air quality and noise issues, along with the frustration of AI and its threat to jobs, are paving the way for a mass people's uprising against big tech. Whether you see this as overdue or not, this moment has the potential to be revolutionary, or at least remarkably disruptive to big tech oligarchs.

This movement isn't quiet; it's gaining traction as the weeks pass and we edge closer to Anthropic and OpenAI IPOs. While I can't speak to these platforms' user numbers, I can speak to this movement growing across social media platforms and becoming something of a united battle on the left. I hope these billionaires, who spend way too much time on social media, are beginning to see this - at the end of the day, this movement is built on anger and rage, the very feelings they exploit for their and their friends' profits. If they are beginning to see pushback against them and their tools, as they race to impress investors and the wider market, then this sudden agreement that their own tools are dangerous to humanity reads like the last cry for public support.

In August, one of the first Billionaires to hold their hands up and say 'this is moving too fast' was Microsoft's Bill Gates, the 'good billionaire' who really wants you to forget what was published about him in the Epstein files. People, at least on social media, responded well to his comments; he gained brownie points and support. But should we listen to Gates? What is his motive here? And what does he gain?

Now, an ex-Anthropic employee has, like many men before him, worked on an LLM, lost his job, and taken to X/Twitter to "warn" the public about the potential harms of this technology. Only this time, instead of the technomanosphere laughing and having its own discourse on X, Reddit, and Discord, the men in charge said, 'yeah, we concur'.

It all sounds like a desperate attempt to win back public support with the promise of guardrails, guidelines, and documents that sound as binding and moral as their Community Guidelines, which quickly switch to harming people at the flick of a late-night Musk tweet.  

Wouldn't it be so awfully convenient to separate yourself from the evil data centre?

We should question everything these remarkably rich men say. Especially when trusting them on this separates the LMM company from the data centre company, making them look like the responsible, concerned guy who has your back. He doesn't; he's a Billionaire.