AI Clones, Steam Discovery and What Valve Told Indie Developers: MyTechDailyNews General Briefing

Knock-off products stopped being a gadget-and-fashion problem years ago. In games, the copycat now has an accelerator, and one studio's warning made that very clear in late October: the imitation can reach shoppers before the original does.

That fear sat at the centre of a long question-and-answer session, where the people who run a major PC storefront explained, in plain terms, who decides what players see and what the platform will not police for them.

The Briefing, in Short

The event was Steam Indie Connect, held on October 27 and co-hosted by Valve, the Busan Indie Connect Festival Organizing Committee and the Korea Creative Content Agency (KOCCA), with developers in Korea as the audience. Tom Giardino and Marc Wilkinson spoke for Valve's Steam Business Team. Matt, from the localization team that handles 30 languages for Steam and Valve games, took the translation questions.

The Threat That Silenced the Room

The moment everyone repeated afterwards can be summed up in one phrase: We'll Copy Your Game with AI and Launch First.

It came from a developer describing a demo recently removed because of cloning problems, in an incident credited to YNOT, along with extortion emails that read, "If you don't pay, we'll clone your game with AI and release it first." The studio argued that Valve's usual advice to show work early could collide with exactly this kind of intimidation, and asked what protection exists.

Tom Giardino said he regretted hearing about such extortion for the first time. He then restated how Valve treats new tooling: "We don't want to be gatekeepers telling developers which tools they should or shouldn't use," which leaves it to each studio to decide whether AI belongs in its process.

Clones themselves, Giardino added, are not new. A striking concept shown at Next Fest can be imitated quickly, trend chasing has long existed, and AI mainly speeds that cycle.

Valve's Two Defenses: People and Momentum

The first defense is community. A developer site, direct player contact, an identity and earned trust blunt clone pressure — though building that takes time and cannot simply be assumed. The second is continuous improvement, and Giardino noted the same worry applies not only to small indie studios but also to Valve, which operates Dota.

He drew a parallel to Valve's anti-piracy approach: instead of aggressive DRM, the company worked to make the legitimate version the best version, and he argued AI cloning deserves the same answer. "Rather than relying on a single mechanic that someone can easily copy, you need to keep growing your game by adding features, personality, and creativity that stem from your own perspective."

Giardino separated cloning from extortion, calling the second mainly a law-enforcement matter — while leaving a door open: "If you report such cases to us, we want to look into them further and help in whatever way we can."

How Big the Platform Has Grown

Before answering, Giardino refreshed the numbers. Steam opened on September 12, 2003 and recently peaked at 43.5 million concurrent users worldwide. In South Korea, over the past five years total Steam revenue grew 126%, premium game and DLC revenue rose 70%, and in-game purchase revenue jumped 388%.

A Launch Is a Starting Gun, Not a Ribbon

The recommended runway, step by step: Coming Soon pages, demos, Next Fest, playtests, an official release date, then build reviews. In Giardino's words, "Launch feels like the finish line, but it's actually the starting line." Attendees later shortened that to another line: Launch Is Not the Finish Line, It's the Starting Line.

He also urged teams to widen reach through consoles, other PC stores, Discord and social media rather than leaning on Steam alone.

Wilkinson covered newer tooling. The Wishlist Data API lets a developer pull wishlist data by period, country and language using an App ID and date range. The Personal Calendar, formerly a Steam Labs experiment and now a store feature, suggests upcoming releases wishlisted by players with similar play styles — and a title appears only after a release date is fixed.

Why Low-Quality Games Surface at Next Fest

Another developer raised AI slop, saying community channels and social media were complaining about low-quality AI output at Next Fest while their own performance felt underwhelming, and asked whether Valve recognized the problem.

Wilkinson explained the visibility mechanics: the first two days rely on virtually random exposure, after which personalized recommendations surface games a player may want to buy. Many complaints about low-quality floods, he noted, came from judging only that opening window. The random start exists to stop abuse from developers entering early just for immediate placement and to give every game a fair chance. "We recognize that the trade-off is some low-quality games get exposed," he said, adding, "We are fully aware of the situation, and a significant number of people are thinking about how to solve this problem."

Giardino remembered the opposite scale. Just three years ago he was urging developers to join with, "There are only 50 or 100 games, so you should definitely join," while participation now reaches into thousands — a reversal he called "the pendulum swinging to the opposite side."

Success, he warned, stays personal. Some studios celebrate 10,000 wishlists because that triples their previous number, while others are disappointed at 50,000 because expectations ran higher. "As platform operators, we can't define what success means for an individual game," Giardino said, noting Valve still opens Next Fest and seasonal sales while improving layouts and placements.

The anxiety has a history, too. Around 14 years ago, when Steam Greenlight began, developers asked whether Greenlight titles could be filtered; when Steam Direct made self-publishing routine, they wanted a checkbox to hide Steam Direct games. Valve has repeatedly chosen to improve discovery rather than exclude developers or prescreen quality. "We don't think we're perfect and there's plenty of room for improvement," Giardino said, while holding that "Rather than limiting developer access, making Steam a smarter and more dynamic platform is the starting point of our philosophy."

Wishlists: Dashboard, Not Trophy

An analysis credited to Chris Zukowski found that sales generated per single Steam wishlist have fallen by roughly half in recent years. That framed the next question: if Steam's front page often centers on top new releases, why shouldn't studios chase wishlist farming — and how do you prepare for personalized placement?

Wilkinson pushed back on the premise. Wishlists show marketing strength, he said, but do not determine placement. "What drives visibility is the number of people buying and playing the game," he said, citing titles that launched with 3 million wishlists and then saw falling sales and dropped off top charts because of poor user experience, alongside games with virtually no wishlists that became massive successes. His advice was to skip farming and make a great game.

Giardino reframed wishlists as measurement rather than a target: check whether they rise after ads, after posting TikTok videos for a week, or after trying other platforms, and remember that methods which helped a first game may fail for a second one years later. "Don't treat wishlists as a magic number you have to hit; use them as a thermometer or speedometer to measure the effectiveness of your efforts," he said.

On the Personal Calendar, he explained the design intent. During Steam Labs only part of users tested it; its official rollout should collect far more response data. It is unusual on Steam because it revolves around recency — top recommendations and top seller lists can still surface older games like GTA 5 or Tactical Breach Wizards, while the Personal Calendar is mainly for brand-new releases. Its limits are real: once 1-2 weeks pass after launch, that exposure stops, and discounts, DLC and game improvements become the ongoing work. Still, "It doesn't matter if this game is a global top seller. What matters is whether there are enough signals indicating it will be popular with you personally," he said, promising further refinement so non-top sellers can still meet their audience.

Tags, Solo Play and the Left 4 Dead Lesson

A developer making a game for two or more players asked whether single-player tags or solo options strongly affect wishlist conversion and refunds, and how to present the store page if the game stays co-op only.

Wilkinson's answer: specific genres or features guarantee nothing. Steam's audience includes VR fans, single-player players and multiplayer players, and co-op-only games like It Takes Two have achieved massive success. Quality decides.

Giardino noted that big releases such as Marathon and ARC Raiders face the same split between solo players and friends, and that Valve has never analyzed wishlist conversion by genre or game type, so he could not claim a single-player tag lifts conversion. Instead he pointed to Valve's own Left 4 Dead: best with three friends, but because not everyone has friends available, the design goal was an experience close to multiplayer while playing solo. If a studio has the budget and resources, build bot companions; if not, step away confidently. "In the end, it's a design choice, not a business choice," he said, adding that a single-player checkbox is not required for success.

For tag selection, he suggested studying competing titles, whose tags, supported languages and features are public. For a game like XCOM or STAR WARS Zero Company, look beyond those two titles and at the games directly beneath them, because broad tags such as Action, Indie or Strategy are shared by around 100,000 games, while narrower ones reveal identity. He also recommended simply asking rival developers: did Brazilian Portuguese support work, which localization agency was used, can you cross-promote through collaborative bundles or recommend each other's games to your fanbases. He found the global indie developer community surprisingly open, and named YouTube reviewers as another route to players.

When 86% of Your Sales Come From Home

A studio with a co-op multiplayer game in Early Access reported growing sales — yet 86% of total sales came from South Korea. Influencers in Japan, Germany and the U.S. were engaged, but Steam's native features after launch did not easily drive overseas sales.

Asked about language, the developer said the store page had been translated into every language supported by Steam and that in-game text had completed LQA for 16 to 17 languages beyond Korean.

Giardino called language a major barrier but noted that some content can perform strongly in particular countries without local-language support, such as farming simulators selling heavily in Germany or Poland — rare cases, in his view, since enjoyable games usually find love across regions. Wilkinson added the pricing angle: developers set prices, but Steam offers a recommended pricing system for teams that cannot analyze pricing individually, and local-market pricing matters for success.

"Breaking out of your home country's fanbase is inherently difficult. Your team isn't doing anything wrong," Giardino reassured, noting Brazilian studios often see most customers are Brazilian too. Titles that win abroad usually have a cheer squad in the new region — local streamers, journalists and YouTube reviewers, not PR agencies or paid ads — so his advice was to find local champions in markets like Germany, the U.S., Canada, Australia, France and China. He also asked for localized update announcements, since games with many languages often share Steam events or feature news only in English, losing Korean and Japanese engagement.

Matt said translating the store page first was correct. Localization, he argued, is a chicken-and-egg problem: you cannot tell whether customers are absent because translation has not happened or whether translation would create them, and the financial risk is real. He recommended using the Wishlist Data API to gather data before deciding, cited recent changes that allow 30 languages to be added at once using AI translation, and noted that conditions differ by region — including Japan, Thailand and Indonesia — so teams should test whether a genre resonates before committing launch plans and resources. He also noted developers in Korea facing the same trade-off.

Steamworks Permissions and Publishers

A developer raised a practical worry: indie teams often need publisher help with marketing and localization, but because Steamworks is tied to accounts, unnecessary or personal data can become visible to publishers. The proposal was to let developers grant permissions only within a contract's scope.

Giardino replied that sharing sales-data viewing and App ID access already exists, but because each contract creates different relationships and permission holders, Valve cannot make it simple for every case. Developer education, he said, is a major issue, since the same requests keep coming at events even for long-available permissions. Valve wants feedback and will improve when specific permission problems are raised.

Steam Frame and VR's Waiting Game

A VR studio CEO asked whether curation, visibility and VR policies would shift with Valve's new hardware, Steam Frame, and voiced a fear common among fans: that Valve might be abandoning virtual reality.

"We still believe in virtual reality," Giardino answered. He revealed that many Steam Frame users play standard library games instead of VR titles and treat the device as a display. VR's problem, he said, is another chicken-and-egg cycle — developers wait for a bigger installed base, players wait for more games, and non-VR titles keep growing more diverse and compelling, consuming players' time and money. He identified new releases like Escape from Tarkov launching every week as the biggest hurdle facing VR developers.

Valve sees Steam Frame as one component of a broader VR ecosystem, and invested in Steam Link for Meta Quest and OpenXR support so users with headsets from other companies can still reach the SteamVR experience. On hardware costs, he noted that higher RAM and memory prices raised expenses for all hardware, including Valve's, even though Valve wanted to price Steam Frame lower to bring in new users. "Moments where we could have welcomed many new customers became reasons to hesitate or wait on VR," he said, adding, "We feel sorry to VR developers."

Checklist: Five Habits That Beat a Copycat

• Own a channel the platform does not control: a developer site, direct player contact, and an identity players recognize.
• Ship on a rhythm. Updates add features and personality that cannot simply be lifted, which is why launch is treated as the starting line.
• Treat wishlists as a thermometer for marketing tests — ads, short-video posts for a week, other platforms — never as a goal to farm.
• Compete on legitimacy rather than lock-in: make the version players buy from you the best version, as Valve argued against aggressive DRM.
• Grow abroad through local champions — streamers, journalists and YouTube reviewers — and localize announcements, not just the store page.

Common Misconceptions, Corrected

What many assume	What Valve said at Steam Indie Connect
More wishlists mean a better store position	They show marketing strength; "What drives visibility is the number of people buying and playing the game,"
Next Fest buries quality games on purpose	The first two days are deliberately random to prevent early-entry abuse; personalization follows
Steam will block AI-made games	Valve does not want to be gatekeepers over which tools developers use
The Personal Calendar keeps you visible indefinitely	That exposure stops once 1-2 weeks pass after launch; discounts, DLC and improvements take over
A single-player checkbox is required to sell well	Co-op-only games like It Takes Two have succeeded hugely; "In the end, it's a design choice, not a business choice,"
VR is being shelved by Valve	"We still believe in virtual reality," with Steam Link, OpenXR and Steam Frame as pieces of one ecosystem

FAQ: What Readers Usually Ask

Is Valve banning AI in games?

No. Giardino repeated that "We don't want to be gatekeepers telling developers which tools they should or shouldn't use," so AI remains each developer's own decision; cloning and low-quality output are discovery problems Valve says it is actively studying.

If someone threatens to clone my game with AI, is that Steam's problem?

Extortion is mainly a law-enforcement matter, but Valve invited reports: "If you report such cases to us, we want to look into them further and help in whatever way we can."

Do wishlists still matter?

As measurement. Sales per single wishlist have fallen roughly half according to the analysis credited to Chris Zukowski, and titles with 3 million wishlists have dropped off top charts while virtually wishlist-free games succeeded. Use them to test campaigns, not as a target.

Should I add a solo mode to a multiplayer game?

Only if your budget supports it. Giardino's example was Left 4 Dead — best with three friends, but designed so solo play feels close to multiplayer — and he framed the outcome as a design question rather than a sales question: "In the end, it's a design choice, not a business choice,"

Why does the Personal Calendar not show my game?

It only lists brand-new releases, and only once a release date is fixed, with exposure ending 1-2 weeks after launch.

How do I sell outside my home country?

Find local champions rather than buying reach, use Steam's recommended regional pricing, and translate update announcements, not just store pages.

Answers, Not Guarantees

The value of the session was that a platform operator explained its own machinery in public. The Q&A confirmed points developers had guessed at: random exposure during the first two days of Next Fest, wishlists not moving visibility rankings, and Personal Calendar exposure limited to 1-2 weeks after launch. Wilkinson also counted how many misconceptions surround wishlists. What Valve did not provide was a final fix for AI slop or cloning — low-quality Next Fest exposure stayed an acknowledged, actively considered problem — but it did clarify what it will and will not do, and it redirected developers toward what they can shape.

For anyone outside the industry, the pattern is worth noticing: when copying gets cheaper and faster, the durable advantage is the relationship — a recognizable identity, a steady stream of updates, and a version worth buying. That is the defence any online product can build, and MyTechDailyNews will keep tracking how Steam's answers on cloning, discovery and VR evolve from here.