Innovation: AI Journals, Cancer Breakthroughs and Data Centre Revolt

Editorial digest April 12, 2026
Last updated : 19:50

Something is shifting in the way we relate to machines — and in the way machines relate to us. This week, three stories from very different corners of the innovation landscape point to the same underlying tension: technology is moving fast, but trust is lagging behind. Whether it's whispering your anxieties to an AI diary, watching a pharma giant report striking trial results, or joining a protest against a data centre being built next to your school, the question is the same. How much of our lives are we prepared to hand over — and on whose terms?

Can an AI diary really become your best friend?

The Guardian's experience piece on AI journalling apps like Mindsera and Rosebud strikes a nerve. The premise is simple: instead of writing into a void, your diary writes back. It comments, suggests, nudges. According to the piece, the author — a lifelong diarist — spent two months using Mindsera and came away describing the experience as making "a new best friend."

That phrase should make us pause. Not because the technology doesn't work — clearly, for some users, it does — but because it raises a question the app developers would rather you didn't ask: what happens to the most intimate thoughts you've ever committed to writing when they live on someone else's server?

AI journalling sits at a peculiar intersection. It borrows the therapeutic language of mindfulness and self-reflection, wraps it in a slick interface, and monetises the result. The appeal is obvious: many people struggle to maintain a journalling habit, and having an AI respond with prompts or reflections can lower the barrier. But a diary that talks back is no longer a diary. It's a relationship — one where the other party has perfect recall, no emotional stakes, and a business model.

For a British audience already wary of tech overreach after years of data scandals, the question isn't whether AI journalling is useful. It's whether "useful" is enough.

Is GSK's Mo-Rez a genuine breakthrough in gynaecological cancer?

Amid the noise, one story deserves more attention than it will probably get. GSK has reported early-stage trial results for Mocertatug Rezetecan — Mo-Rez — a treatment targeting ovarian and endometrial cancers. The numbers, according to GSK, are striking: tumours shrank or disappeared in 62% of ovarian cancer patients where chemotherapy had already failed, and in 67% of those with endometrial cancer.

A word of caution: these are early-stage results. The trial population is small. Pharma companies have a long history of announcing impressive preliminary data that softens in phase three. GSK's CEO Luke Miels is reportedly pushing to accelerate the company's drug development pipeline, which adds a commercial incentive to the timing of this announcement.

But even with those caveats, the clinical need here is enormous. Ovarian cancer remains one of the deadliest forms of the disease, in large part because it is typically diagnosed late and responds poorly to existing treatments once chemotherapy fails. If Mo-Rez holds up through larger trials, it could become a significant weapon in a therapeutic area that desperately needs one.

For the UK specifically — where NHS cancer waiting times remain a politically charged issue — a viable new treatment from a British-headquartered pharma company would carry weight well beyond the lab.

Why are Americans revolting against data centres?

Perhaps the most politically significant story this week is the growing backlash against AI data centre construction across the United States. As The Guardian reports, opposition now spans the political spectrum: Republican Texas, Maga-friendly midwest states, Bernie Sanders, and Californian teachers' unions are all pushing back against what they see as an unregulated building boom.

This matters for the UK because the same pressure is coming. The British government has signalled its enthusiasm for AI infrastructure. Planning reforms are being pitched partly as a way to fast-track data centre approvals. But the American experience shows what happens when communities feel the costs — noise, water consumption, strain on power grids — land on them while the profits flow to Silicon Valley.

The White House had made rapid data centre rollout a centrepiece of its AI strategy. That plan is now meeting democratic friction. When voters in deep-red Texas districts and progressive Californian towns agree that something has gone wrong, politicians tend to notice.

The lesson for Westminster is straightforward: if you want public consent for AI infrastructure, you need to offer more than jobs and GDP projections. Communities want environmental safeguards, genuine consultation, and a share of the value being created. Skip that step, and you get what America is getting — a revolt that no amount of techno-optimism can talk away.

What ties these stories together?

A cancer drug trial, an AI diary, a political backlash against data centres — on the surface, these have little in common. But each one exposes the gap between what technology promises and what people actually experience. GSK promises hope, but the road from early trial to NHS prescription is long and littered with disappointments. AI journalling promises self-knowledge, but on terms set by a company you've never met. Data centres promise economic transformation, but the communities living next to them see rising electricity bills and falling water tables.

Innovation doesn't fail because the technology doesn't work. It fails when the people it's supposed to serve stop believing the deal is fair.