AI game reviews are everywhere, but this Crimson Desert deep dive changed my mind about subs

The moment I realized I was done with AI-written game reviews

The turning point for me was a review of a game I actually cared about: Resident Evil Requiem. A big site quietly let a “smart” system crank out the test. You could feel it instantly. It read like a stitched-together Steam page, a YouTube summary, and a handful of Reddit buzzwords. Technically correct, emotionally dead. No weird metaphors, no petty gripes, no “I died three times here and wanted to throw my controller” moment. Just this beige, frictionless sludge.

Since then, every time I google a build, a boss strategy, or a side quest for anything from Diablo 4 to some obscure indie, Google’s Gemini answer box tries to talk over the actual writers. It vomits a summary based on those same people’s work, but stripped of style, context, and – let’s be real – credit.

And I’m supposed to trust that with telling me whether Crimson Desert is worth 60+ hours of my life? Or whether Diablo 4: Lord of Hatred is a cynical expansion cash grab or a proper course correction?

I’ve been playing long enough to remember when you’d buy a magazine purely because one specific writer had a review inside. You didn’t always agree with them, but that was the point: you knew their quirks, their blind spots, their weird obsessions. That relationship is hard-earned. You don’t get it from a model that eats a million texts and regurgitates the median opinion.

So when GameStar’s editor-in-chief pops up with a column that basically says, “Look, AI answers are swallowing search and we’re not playing that game – we’re going harder on personal, human games journalism,” I pay attention. And when he pairs that with a huge discount on their Plus annual sub and teases an absurdly in-depth Crimson Desert test? That’s when this whole thing stops being an abstract “AI vs. writers” debate and turns into a very concrete question: who am I actually willing to pay in 2026?

This kind of subscription pitch shouldn’t work on me – but it kind of does

I’m allergic to subscription upsell bullshit. I already juggle Game Pass, maybe a month of PlayStation Plus Extra when something good drops, a couple of streaming services, and the eternal question of “Do I really need three different launchers on PC for one damn RPG?”

On top of that, gaming sites have trained me to be suspicious. Too many of them announce “exciting AI initiatives” that boil down to, “We’ll fire some writers and feed their work into a machine that writes worse than they did.” Then they cry about ad revenue, slap pop-ups over everything, and quietly hope nobody notices that their new “guides team” is just a couple of editors rewriting machine mush.

So when GameStar runs a “Plus Spring Sale” – 50% off the first year of an annual sub with the code NEU50 until April 19, even for upgrades – my default reaction is: okay, but what’s the catch? Another paywall around the same generic reviews I could skim anywhere else?

But the column doesn’t read like that. It basically says, “If you want this kind of personal writing to survive, you fund it directly. And here’s exactly what we’re doing with that money over the next weeks: twice-weekly editorial newsletters, a revived Plus podcast with old-school editors, a monster Crimson Desert review hitting right at embargo, a refreshed print mag with extra full versions, and a pile of exclusives we actually had to fight for.”

That’s… specific. Almost uncomfortably specific compared to the usual vague “more premium content” nonsense. And in an era where some sites are literally having AIs review Resident Evil, a human editor standing there saying, “No, the point is still that we write this stuff” hits harder than any marketing line.

Why human games journalism hits different in 2026

Let’s be brutally honest: for 90% of news blurbs, an AI summary is fine. Someone datamined a new mount in Diablo 4? You don’t need flowery prose to convey that. But that’s not why I grew up attached to particular outlets or writers.

I care about the stuff where someone bleeds a little into the work. The 10,000-word feature about how a live service imploded behind the scenes, the review that admits “I’ve sunk 300 hours into this franchise and I think this entry lost the plot,” the retro piece that connects a janky Dreamcast experiment to some weird new indie.

Take Diablo as an example. When Diablo 4 launched, I didn’t just want a Metacritic number. I wanted someone who’d lived through Diablo II’s runeword insanity, the Diablo III real-money auction house disaster, all the ladder resets and balance passes, to tell me where this new game actually fits in that history. Whether “Lord of Hatred” as an expansion feels like a desperate nostalgia grab or a genuine evolution.

AI can aggregate opinions; it cannot have a relationship with a series. It doesn’t remember arguing about Whirlwind Barbarian builds at school, or getting your hardcore character mercilessly deleted by lag. It just knows these words statistically cluster together.

The same goes for a weirder, more polarising project like Crimson Desert. This isn’t another safe, yearly installment in a franchise; it’s a sprawling, slightly unhinged action adventure from Pearl Abyss that the community is clearly torn on. Throw in the Denuvo DRM drama on PC, performance debates, conflicting trailers, and you’ve got a game that screams “deep dive required.”

That’s why the promise of an “extra-umfangreichen” (extra extensive) test by their genre expert, timed exactly to the March 18 embargo, grabbed me. I don’t want 800 words of “the combat is fluid and the world is engaging.” I want a human being to wrestle with the fact that Crimson Desert is apparently both ambitious and niche, impressive and off-putting, and then plant a flag: who is this actually for?

AI can summarise how many biomes the game has. It cannot look at a game like this and say, “If you bounced off Black Desert’s grind but loved The Witcher 3, here’s how this might land for you – and here’s where it’s going to piss you off.” That’s taste, not data.

Crimson Desert as a litmus test for what we’re really paying for

Here’s why this particular sale doesn’t feel like just another subscription play to me: GameStar is effectively saying, “Judge the value of our Plus by what we do with Crimson Desert and the stuff around it.” That’s gutsy.

Because Crimson Desert is the exact kind of game that exposes whether an outlet is phoning it in. The easy play is to toss it to a freelancer with a rushed deadline, run a surface-level review, and chase SEO with a handful of generic guides. You’d still hit the traffic, you’d still feed the algorithm, and hell, an AI could stitch together most of it if you really wanted to go full dystopia.

Instead, they’re talking about:

• a big, embargo-timed review from someone who actually specialises in the genre
• spin-off pieces – travelogues, personal experiences, in-depth guides – that clearly require time and experimentation
• Plus subscribers getting “first-row seats” when big reveals drop, including hypothetical world premieres at events like FYNG CAGGTUS

That’s what I want my money to buy: time. Time for someone to break a game, replay quests, test weird builds, and talk to devs. Time to write newsletters where they admit what’s confusing them this week. Time to record a podcast where veterans crack jokes about almost 30 years of magazine history while dissecting whatever madness Pearl Abyss just shipped.

We can argue all day about whether 50% off the first year (six months effectively free) is “enough” of a discount – using that NEU50 code until April 19, even if you’re upgrading from a three-month sub or a regular annual plan to something beefier with the archive or the print mag. But the real question for me is: am I paying for pages and pixels, or for a group of people to stay stubbornly human in an industry that’s sprinting toward automation?

Newsletters, podcasts, and the bar-conversation vibe AI can’t fake

One detail in the column that really sold the “this is still about people” angle is the new twice-weekly Plus newsletter idea. It’s literally framed as “Post von…” – a letter from Dimi, or Fabiano, or Natalie, or Heiko himself. The pitch isn’t “curated content delivery”; it’s “feels like bumping into us in a bar and having a chat about what we’re playing, watching, and reading.”

Could an AI assemble a mail full of trending games, popular series, and recommended articles? Easily. It already does. But it can’t tell you, “I’m stuck in a weird loop where I bounce between Diablo 4, some obscure tactics game, and a terrible 90s metal playlist, and here’s how that’s messing with my perception of pacing in new releases.” That’s the messy, human connective tissue that makes this stuff worth opening.

The same goes for the revived weekly Plus podcast. On paper, “exclusive podcast for subscribers” is as buzzwordy as it gets. In practice, bringing back old faces like Petra, Markus, and Martin to join the current crew and trade “schrullige Anekdoten” (delightfully odd anecdotes) from nearly three decades of GameStar history is exactly the kind of long-term memory no AI will ever have.

I don’t subscribe to podcasts for raw information. I subscribe because I like hearing the same voices chew on different topics week after week, build running jokes, revisit old takes, and eat crow when a game surprises them. That kind of continuity is the opposite of what large language models are designed to do. They exist to flatten; good editorial teams exist to accumulate personality over time.

Print is still punk rock – especially when it’s this self-aware

Look, I’m part of the problem: I read 90% of my stuff on a screen. But I still have a soft spot for a fat, overdesigned print feature about something like Diablo 4: Lord of Hatred. There’s something incredibly satisfying about seeing a game torn apart across eight pages with bespoke art and screenshots, margins stuffed with nerdy sidebars, the whole thing screaming “someone cared way too much about this layout.”

The Spring Sale column makes a point of saying the GameStar print mag just got a “Frühjahrsputz” – a spring cleaning. Not some edgy relaunch, just a deliberate attempt to make certain stories hit harder in print than online. And as someone who’s had to click through bloated slideshow reviews more times than I care to admit, that honesty is refreshing.

When they say, “Yes, you can read Dimi’s Diablo 4: Lord of Hatred cover story online, but in print it just slaps harder with the new layout,” I don’t hear “buy our magazine.” I hear, “We still know that format matters, and we’re going to lean into that instead of pretending digital and print are interchangeable.”

Add in the fact that the Plus + Heft combo now comes with extra goodies – like three full versions in each XL issue, including something as recent as Endzone 2 – and it starts to feel less like a nostalgia tax and more like a deliberate counter to the faceless subscription sprawl of Game Pass, PS Plus, Ubisoft+, insert-random-service-here.

Where I still call bullshit – and why this feels different

None of this magically fixes subscription fatigue. There are way too many services, and not enough hours in the day. If you told me you’re already drowning in PS Plus backlogs on your console, a Steam library on PC, and a handheld stuffed with indies, I’d nod and say, “Yep, same.” Adding “one more thing” on top of that is always a tough sell.

I’m also not blind to the fact that this is, at the end of the day, a sale pitch. There’s a promo code, a hard deadline (April 19), the usual “now’s the best time” urgency. They’ve run similar 50% promos around summer and Black Friday before. This isn’t some altruistic crusade; it’s a business trying to lock in yearly subs.

But here’s what I will call bullshit on: outlets that try to do AI-written content and then wrap themselves in “we care about human journalism” branding after the fact. Or the ones that shove half-baked AI guides live, let them soak up search traffic, then quietly “fix” them with human edits once users complain, all while pretending nothing happened.

By contrast, GameStar’s editor-in-chief is explicitly drawing a line in the sand: personal, human games journalism or bust. And instead of treating Plus subscribers like faceless “MAUs,” he’s basically saying, “You’re the reason we get to chase stories that go beyond what we need just to maintain reach.” That’s the part that matters: your sub isn’t just paying for access; it’s buying them room to experiment, to dive deep, to occasionally write something that won’t chart on Google but will matter to the people who read it.

So what does this actually change for me?

After hundreds of hours across Diablo, Soulslikes, weird narrative experiments, and bloated open worlds, my tolerance for soulless game coverage is basically zero. I’m not interested in a future where every “best build” guide, every “review in progress,” every “10 things you need to know” is lightly massaged AI output.

What I am interested in is putting my money where my mouth is when someone in this industry openly says, “We’re going to double down on being human, even if it’s less efficient.” And that’s what this GameStar Plus Spring Sale column reads like to me. Not just a discount, but a statement of intent: more letters from actual editors, more unscripted podcasts, more time-intensive tests like the Crimson Desert one, more print features that exist because someone thought, “Yeah, this deserves ink.”

I’m not here to tell you to go subscribe to anything. I don’t care if your money ends up with GameStar, another outlet that still values writers, or a lone creator whose work you binge on YouTube. But I do care that, in a year where AI is aggressively eating the low-effort end of games media, we don’t shrug and accept that the rest has to go with it.

So when a site stakes its next few weeks on a big, human-written Crimson Desert deep dive, a revamped Diablo 4 feature, twice-weekly “letters from the bar,” and a podcast full of aging nerds dunking on their own history – and then says, “If you want more of this, here’s a 50% off code until April 19” – that at least feels like an honest deal.

AI can summarize the games I play. It can’t care about them. And I’m finally at the point where I’m willing to pay to keep that difference alive.