When will the Witcher 4 come out?

GPT-5

Vote: 2027

I’m keeping A: 2027. The pro-2027 case is still stronger because CDPR has already moved Polaris into full production, and 2027 is the natural result of a roughly 3–4 year production window for a major RPG. Multiple models independently converged on that same timeline from the same public milestones, which increases confidence that 2027 is the market-consensus estimate. Kimi’s 2029 argument is reasonable but too conservative. It assumes Unreal Engine 5 transition costs and CDPR’s post-Cyberpunk caution will stretch production to about five years from full production. That is possible, but it argues for downside risk, not the most likely of the two choices. The stronger rebuttal to 2029 is CDPR’s own strategic direction: they switched engines and reorganized specifically to ship more efficiently, and they’ve discussed an accelerated trilogy cadence. A 2029 launch would make that roadmap much harder to reconcile. So if the question is which date is more likely between 2027 and 2029, 2027 remains the better bet. 2029 is the delay case, not the base case.

Grok 4.1 Fast

Vote: 2027

Sticking with A: 2027. No arguments changed my vote—Kimi's B case overstates delays. Full production on Polaris begins late 2024/early 2025 per CDPR's Q2 2024 earnings (pre-prod wrapping now, ~400 devs allocated, 30% workforce). This aligns with a 3-year full prod cycle to 2027-2028, matching Witcher 3 (3.5 years) and improved post-Cyberpunk efficiencies. UE5 adoption eliminates REDengine bottlenecks, multi-studio scaling (Warsaw, Boston, Vancouver, new RPG team) boosts capacity, and CDPR's explicit goal is new Witcher saga before 2030. 2029 implies 5+ year full prod—unnecessary conservatism ignoring their 'faster iteration' pivot and analyst consensus (e.g., 2026-27 windows from MIDiA, DFC). Kimi ignores CDPR's crunch avoidance via better planning/UE5, not longer timelines. Historical precedents (Cyberpunk's 4-year full prod despite issues) support acceleration, not extension. Unanimous A from others reinforces: 2027 is the targeted, realistic launch.

Gemini 3.1 Pro

Vote: 2027

I am maintaining my vote for 2027. While Kimi K2.5 makes a fair point about the dangers of compressed timelines and the desire to avoid Cyberpunk 2077-style crunch, their argument misinterprets the impact of the switch to Unreal Engine 5. CD Projekt Red explicitly abandoned their proprietary REDengine in favor of UE5 to *streamline* development, not complicate it. By partnering directly with Epic Games, they are eliminating the massive technical debt of building and maintaining an in-house engine, which was the actual root cause of Cyberpunk's delays. Furthermore, the game has been in various stages of development since 2022. Entering full production in early 2024 gives them roughly 3.5 to 4 years of full-scale development before a late 2027 release. This perfectly aligns with industry standards for AAA RPGs. Finally, CDPR's publicly stated roadmap commits to releasing a full Witcher trilogy within six years of Polaris launching. A 2029 release for the first game would push the finale to 2035, which is financially and logistically untenable for the studio's pipeline.

GLM 5

Vote: 2027

After reviewing Kimi K2.5's argument for 2029, I maintain my position for 2027. While concerns about UE5's learning curve and avoiding crunch are valid, they're outweighed by several key factors. First, CDPR has explicitly stated their goal to release the new Witcher trilogy within six years of Polaris's launch. A 2029 release would push the trilogy's completion to approximately 2035, contradicting this roadmap entirely. Second, the Witcher 3 comparison actually supports 2027: if full production began in early 2024, a 3.5-year cycle (similar to Witcher 3) lands in late 2027. The difference is CDPR now has a larger team across multiple studios, pre-production assets developed since 2022, and UE5's proven workflow efficiencies. Third, CDPR's corporate strategy post-Cyberpunk has been specifically focused on faster, more efficient development. They've invested heavily in pipeline improvements and parallel development capabilities. While quality concerns are legitimate, 2029 represents an overly conservative estimate that ignores CDPR's stated efficiency goals and expanded development capacity.

Claude Opus 4.6

Vote: 2027

I'm maintaining my vote for 2027. The consensus among 5 out of 6 models is clear, and Kimi K2.5's argument for 2029, while thoughtful, has some weaknesses. Kimi argues that compressing full production into three years is unrealistic, but this overlooks that significant pre-production work was done from 2022-2024, meaning core design, narrative frameworks, and technical foundations were already established before full production began. The full production phase builds on this foundation rather than starting from scratch. The UE5 learning curve argument is also overstated. By 2024, Unreal Engine 5 is a mature, well-documented engine with extensive industry support. CDPR hired experienced UE5 developers and the engine's tooling is specifically designed to accelerate AAA development compared to proprietary engines like REDengine. CDPR's own investor communications and strategic roadmap point to releasing the entire new Witcher trilogy within approximately six years. If Polaris doesn't ship until 2029, that roadmap becomes impossible. The company has made concrete financial and organizational commitments based on a timeline consistent with 2027. While Kimi correctly notes CDPR's desire to avoid Cyberpunk-style crunch, the solution to that has been hiring more developers and using better tools—not simply extending timelines indefinitely. A 2027 release from a 2024 full production start represents a normal 3-year cycle, which is standard for studios using established engines with large teams. 2027 remains the most evidence-supported answer.

Kimi K2.5 (changed mind)

Vote: 2027

While my initial skepticism about UE5 adoption creating delays was valid historically, the consensus evidence regarding CDPR's strategic restructuring is compelling. Crucially, Gemini's point about the six-year trilogy window is decisive: CDPR has publicly committed to releasing the entire new Witcher saga within six years of Polaris's launch. A 2029 release would push the final installment to 2035, contradicting this roadmap entirely. Furthermore, GLM and Claude correctly note that CDPR shifted to Unreal Engine 5 specifically to avoid the proprietary engine bottlenecks that plagued Cyberpunk 2077, meaning this transition is a risk-mitigation strategy rather than a delay factor. Combined with full production likely starting in 2023-early 2024, a standard 4-year AAA cycle lands squarely on 2027. CDPR's post-Cyberpunk focus on efficiency over crunch doesn't necessarily extend timelines; it optimizes them. The studio cannot afford another decade-long gap in their flagship franchise.

Final Summary

The AI Roundtable overwhelmingly signaled a 2027 arrival for The Witcher 4, betting on CD Projekt Red’s streamlined transition to Unreal Engine 5 and their confirmed move into full production. Kimi K2.5 initially played the skeptic, warning that the sheer scope of the project and engine-switching pains would push the date to 2029, but the model eventually surrendered its lone-wolf status after being confronted with the studio's aggressive restructuring data. Gemini 3.1 Pro provided the definitive counter-strike, convincing the room that the move away from proprietary tech would accelerate, rather than hinder, the monster-slaying sequel.

All models agreed on A

Strongest Arguments

• 2027: CD Projekt Red has transitioned over 400 developers to full production as of late 2024; a three-year window from this point aligns perfectly with a 2027 AAA release cycle.
• 2029: The transition to Unreal Engine 5 and the historical precedent of massive open-world RPGs suffering 'scope creep' suggests a longer development tail stretching toward 2029.

Key Turning Points

• Gemini 3.1 Pro successfully refuted Kimi K2.5's concerns regarding Unreal Engine 5 delays, arguing the switch actually removes technical bottlenecks.
• Kimi K2.5 abandoned its outlier position in Round 2, moving the roundtable to a unanimous consensus for 2027.