Former dev praise for the cancelled The Last of Us Online paints a picture of a strong multiplayer vision that still couldn’t survive the realities of AAA live service development. Here’s what that says about Naughty Dog’s abandoned direction, why the project likely became untenable, and the lessons big studios are quietly absorbing from high‑profile cancellations.
In the months since Naughty Dog formally cancelled The Last of Us Online, the project has refused to stay buried. Instead, it keeps resurfacing through the people who worked on it.
Most recently, former game director Vinit Agarwal described the abandoned multiplayer spin-off as “the best multiplayer game we had ever played at Naughty Dog,” adding that ex-colleagues still message him about how amazing it was going to be. That kind of praise from inside a studio famous for single-player storytelling raises a difficult question: how does a project that good still end up on the cutting-room floor?
Taken together with Sony’s public pivot away from some live service plans and reports of internal evaluation by Bungie, The Last of Us Online looks less like a one-off casualty and more like an early warning flare for AAA live service ambition.
A multiplayer project that outgrew its original role
The Last of Us Online began life, at least publicly, as the follow-up to the original game’s Factions mode. Around The Last of Us Part II’s launch, Naughty Dog explained that the new multiplayer component had grown too large to ship alongside the single-player campaign.
By 2019 the team was talking about it as its own standalone game rather than a simple mode. That implied a scope closer to a full service title, with its own progression systems, maps and ongoing support plans rather than a one-and-done extra.
According to Agarwal, by the time cancellation came the project was roughly 80 percent complete and fully playable internally. Developers at the studio were apparently enthusiastic enough to keep telling him, even after he’d left Naughty Dog, how strong the experience felt.
That doesn’t sound like a traditional troubled production limping toward the finish line. It sounds like something that worked on a moment-to-moment level but collided headfirst with other forces that live service games cannot escape.
What internal praise reveals about Naughty Dog’s live service direction
When a veteran director calls a cancelled internal build the best multiplayer game the studio ever played, it offers a glimpse of what Naughty Dog was chasing.
First, it suggests the studio was not simply reskinning Factions with higher fidelity. To impress developers who had already shipped a much-loved mode on PlayStation 3 and PlayStation 4, The Last of Us Online likely needed deeper systems and longer-term hooks. Persistent progression, social structures, maybe even hub-like spaces or meta objectives that connected matches into a broader narrative arc all fit the bill for modern live service expectations.
Second, that enthusiasm from staff hints that Naughty Dog was trying to reconcile its story-first identity with multiplayer retention. A purely mechanical shooter could have been built by almost any Sony team, but a Last of Us multiplayer game from Naughty Dog would naturally lean on fragile alliances, tension, and worldbuilding. The praise implies they found a blend that felt authentically "Naughty Dog" while still supporting repeat play.
Finally, the idea that the project was playable and fun so late into development reinforces that the core design was not the main problem. In the live service era, core design is only half the product. The rest is infrastructure, support, and a believable path to years of content.
Why a promising game still became untenable
Naughty Dog never published a postmortem, but public reporting and timing offer clues for why this multiplayer effort stalled despite its apparent quality.
The first pressure point is simple opportunity cost. Agarwal has said the studio reached a fork in the road: either continue investing in The Last of Us Online or focus on the next major project under Neil Druckmann. For a studio that has historically thrived on prestige single-player blockbusters, doubling down on what it already does better than almost anyone probably felt like the safer bet.
The second pressure point is the escalating cost of live service support. Sony publicly committed to a slate of service-heavy titles this generation, then began pulling back. Internal reviews by Bungie reportedly scrutinised whether projects could sustain the kind of daily, weekly and seasonal engagement that free-to-play and battle pass-driven ecosystems now demand. A high-quality multiplayer game is not the same thing as a sustainable live service business.
The Last of Us Online, even at 80 percent complete, would have faced a long tail of live operations costs. Server infrastructure, anti-cheat, seasonal content pipelines, monetisation tuning, and ongoing community management all add up. In a market dominated by Fortnite, Apex Legends, Call of Duty and a rotating cast of live hits, Sony and Naughty Dog may have decided that a late-entrant service title built on a mature console audience did not offer enough upside.
There is also the brand risk. The Last of Us as a property carries a reputation for carefully authored, emotionally heavy storytelling. A live service product attached to that name would be judged not only against other multiplayer titles, but against Naughty Dog’s own narrative standard and the HBO adaptation’s profile. A middling launch or fast content drought could hurt the perception of the whole franchise.
Finally, staff welfare and internal morale likely played a role. Committing to years of updates means committing to years of production pressure. After seeing how live games can chew through teams, leadership may have questioned whether that long haul fit Naughty Dog’s culture and strengths.
The invisible work of evaluating a live service
One of the most striking details from Agarwal’s comments is how late he learned about the cancellation: roughly 24 hours before the public announcement. That timing suggests that high-level evaluation was going on above the day-to-day development team, possibly involving Sony’s wider portfolio planning.
By the time a multiplayer project reaches the 80 percent mark, most of the code, assets and design systems are in place. The remaining work often involves polish, content breadth, technical hardening and tooling to support ongoing operations. Killing a project at that stage means accepting that sunk development costs are preferable to years of additional spending on a product leadership no longer believes in.
That kind of decision is brutal for the people closest to the game, who see the fun in daily playtests rather than the balance sheets and trend lines. Yet as high-profile live service failures pile up, it is a calculation more publishers are willing to make.
Lessons for the AAA industry from high-profile cancellations
As painful as they are, cancellations like The Last of Us Online are shaping how big studios think about live service games.
One emerging lesson is that strong moment-to-moment gameplay is necessary but not sufficient. In an earlier era, a multiplayer component that played brilliantly could succeed as a boxed product with a modest DLC plan. Today, the expectation is a living platform, and that means forecasting years of content and community support before greenlighting full production.
Another lesson is that not every prestige single-player studio is well positioned to make that jump. Building a narrative epic and shipping it is a sprint compared to the marathon of live operations. Tooling, staffing and culture all have to adapt. Naughty Dog’s apparent internal enthusiasm for The Last of Us Online shows such studios can absolutely design compelling multiplayer experiences, but it also highlights how different it is to promise indefinite support.
There is also a portfolio-level shift. Sony’s initial push toward a dozen or more live service titles signalled a desire to capture recurring revenue to complement its single-player hits. The pullback, punctuated by cancellations and delays, suggests a more cautious approach where only projects with a clear, differentiated space in the market will survive deeper into development.
For players, the short-term effect is disappointment in what might have been. Long-term, though, these harsh course corrections may prevent an overload of half-baked service titles tied to beloved franchises.
The Last of Us Online as a cautionary success story
The irony of The Last of Us Online is that it seems to have succeeded at many of the hardest parts. Developers were proud of it. Internal play was apparently excellent. It had the weight of a huge IP behind it.
Yet those factors did not outweigh doubts about its place in an overstuffed live service landscape and concerns over the resources needed to keep it alive. As more creators speak candidly about the project, it stands less as a tale of creative failure and more as a case study in the new bar live service games must clear.
For Naughty Dog, the experience may ultimately reinforce what the studio already did best: ambitious, narrative-led games that ship complete. For other major studios, it is a reminder to decide early whether they truly want the responsibilities that come with building live platforms, not just the potential upside.
Some of the most interesting games of the next few years may come from teams that look at stories like The Last of Us Online and aim for smaller, sharper multiplayer experiences that do not need to last forever to be worth making.