Larian CEO Swen Vincke responds to backlash against AI use while making its new RPG Divinity — "It would be irresponsible for us not to evaluate new technologies"

Larian Studios is facing community backlash after reports it used generative AI for ideation on its upcoming RPG Divinity; CEO Swen Vincke clarified AI is an optional tool for developers to generate reference images, asserted no AI content will ship in the game, and announced an AMA to address questions about the dev process. Vincke framed evaluating new technologies as responsible while noting the studio will change course if the tools conflict with its values, creating reputational and IP/ethical exposure for the studio but posing limited direct market impact given Larian’s private status.

Analysis

Market structure: The immediate winners are AI infrastructure and cloud vendors (NVDA, AMD, MSFT, GOOGL, AMZN) as game studios evaluate optional tooling — expect incremental cloud/GPUs spend of ~3–8% for large studios over 12 months. Losers are reputationally-exposed indie studios and creative marketplaces that rely on user trust; consumer backlash can transiently reduce pre-order conversion by an estimated 1–5% for affected titles over release windows. Pricing power shifts toward vertically integrated tool providers who can offer provenance and licensing guarantees.

Risk assessment: Tail risks include regulatory action (EU/US transparency/compensation rules) or landmark IP litigation that could curtail model use — plausible within 6–18 months and capable of knocking 5–15% off addressable revenues for cloud/AI services in gaming. Short-term (days–weeks) effects are reputational noise; medium-term (3–12 months) is policy and legal clarity; long-term (2–5 years) is structural adoption of non-infringing training pipelines. Hidden dependency: many studios rely on third‑party foundational models; supplier concentration (Nvidia GPUs, Azure/GC/ AWS) is a single-point operational risk.

Trade implications: Direct overweight AI infra and cloud (NVDA, MSFT, GOOGL) with nimble sizing (1–3% positions) while underweight small-cap game developers and marketplaces lacking IP controls. Use pair trades (long NVDA, short a headline-sensitive publisher if it sells off >7%) and options (buying 3–9 month call spreads on NVDA/MSFT to cap premium). Entry: scale in over 2–6 weeks; add on pullbacks >8%; initial stop-loss 8–12%.

Contrarian angles: The market is overestimating near-term consumer rejection — historical parallels (content moderation scares) show revenue rebounds within 3–9 months once policies are clarified. A sustained PR backlash could instead accelerate demand for paid, provenance-ready tooling (a new revenue stream for cloud vendors and middleware firms like UNITY if they productize it). Monitor AMA (weeks) and first lawsuits (3–12 months) as binary events that could create 10–25% repricing opportunities.

Key Decisions for Investors

• Establish a 1–2% portfolio long in NVIDIA (NVDA) over the next 2–6 weeks to play continued GPU demand from studios; add another 0.5–1% on pullbacks >8%; target 12-month horizon, stop-loss 12%.
• Build a 1–2% position in Microsoft (MSFT) over 4–8 weeks to capture Azure AI studio spend; trim if Microsoft Cloud guidance slips by >3% QoQ or if regulatory risks materialize within 90 days.
• Implement a pair trade: long NVDA (0.75–1%) and short Activision Blizzard (ATVI) or Electronic Arts (EA) (0.5–0.75%) only if a headline-driven selloff exceeds 7% intraday; target mean reversion within 3–9 months, take profits at +10–20%.
• Use options for leveraged exposure: buy a 3–9 month NVDA call spread sized to 0.5–1% portfolio risk (buy 30‑delta, sell 60‑delta) to limit premium risk while retaining upside to earnings/cloud demand prints.
• Set a regulatory/litigation trigger: if a US/EU draft bill or a class-action naming major cloud/model vendors is filed/passes committee within 30–180 days, reduce net AI infra exposure by 50% and rotate into defensives (Staples/Utilities) until language and financial impacts are quantified.