A preregistered Ipsos survey of 9,000 respondents across China, Finland, France, Germany, Italy, Spain, Taiwan, the UK, and the US (January 2026) finds that publics are not categorically opposed to military AI applications. Support is highest for non-lethal decision support and falls as lethality and automation increase, with fully autonomous lethal force the least-supported application in every country. The strongest predictor of overall support is general AI benefit perception, not principled opposition to lethal autonomy, meaning that civilian AI discourse and everyday product experiences shape tolerance for military applications more than arms-control framing does. For financial institutions, this suggests that public-opinion constraints on government-AI-vendor military contracts are unlikely to be broad-based, but remain a concentrated and measurable risk specifically for products enabling fully autonomous lethal targeting.
271 other tracked developments touching the same claims.