aiqbench analysis
Good Start Labs demonstrates that fine-tuning frontier models on complex strategy games like Diplomacy can transfer specific capabilities, such as strategic planning and financial research, to real-world benchmarks. This approach leverages verifiable game outcomes to shape AI behavior in customer support and industrial operations, highlighting a novel pathway for improving model utility through gaming environments.
Can Skills Learned in Games Transfer to Real-World Work?
Good Start Labs trained an AI on a railroad game — and one version improved at financial research. The difference was the training design.
“Games have always been these underrated educational tools. They’re super approachable. They’re very human.” Those are the words of Alex Duffy, co-founder and CEO of Good Start Labs, who spoke to Latent Space about why his company is turning games into training material for AI models. The company was spun out of AI media and tools company Every last October, with $3.6 million in funding from General Catalyst, Inovia, Every, and angel investors. The idea came from a 2025 Twitch stream of frontier models playing the game Diplomacy, which Duffy said normally takes “days or weeks to play.” This was when he worked at Every as its head of AI training. 2025 Twitch stream showing a Diplomacy betrayal by OpenAI’s o3 model. (For more on Diplomacy and LLMs, see our interview last year with Noam Brown, soon after he won the 2025 World…
Excerpt shown under the publisher's RSS terms · 143 words · full text and figures are on the publisher's site
Related coverage
Game Design Creativity evals
LessWrong · Oct 12, 2026
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · Sep 25, 2026
Echoverse: Deep, evolving environments for computer-use agents
Microsoft Research · Jul 30, 2026
Dyna-2: General Purpose Physical Agents with Hour-Long Autonomy — Jason Ma, Dyna Robotics
Latent Space · Oct 13, 2026