Aidan Gomez, Co‑Founder and CEO of Cohere, says businesses increasingly care not just about model performance or cost but about who controls their AI infrastructure. After Cohere combined with Germany’s Aleph Alpha in April, the company’s valuation rose from $7 billion a year earlier to $20 billion.
Gomez argues that geopolitical tensions have made firms aware of the risk of getting caught in political crossfire — for example, when the United States imposed export controls on Anthropic’s Mythos model in June. Cohere presents itself as an alternative that can strengthen the resilience of technology supply chains and provide companies with “AI sovereignty,” meaning full control over their AI systems.
"Once a customer deploys Cohere’s software, we can’t see in. We can’t switch it off. We can’t intervene in any way," Gomez said, adding that executives now worry about control as much as about model capability or cost. He believes competing in AI is not merely about building the largest model or spending the most money.
Sceptics call sovereign AI a pipe dream, arguing it would be prohibitively expensive to replicate major US or Chinese technologies. Gomez responded that sovereignty does not require everyone to copy everything — that would be highly inefficient — but companies should diversify suppliers to avoid single points of failure or overreliance on technology from authoritarian states.
Market research points to rising concern: one analysis estimated that up to 40% of AI spending could be influenced by sovereignty requirements by 2030. This week Cohere released an IDC study of executives across four countries showing that most respondents consider AI sovereignty important, but few can clearly define it or have concrete plans to achieve it.
Gomez frames Cohere’s mission as strengthening the sovereignty of democracies adopting AI, focusing on critical sectors such as telecoms, financial services, healthcare, the power grid, and water treatment. He said more democracies need to build AI capabilities and create robust, resilient global supply chains rather than leaving transformational technology to the United States alone.
He also argued that Canada is a strong place to build such a company, citing perceived trustworthiness and shared values as advantages that help a Canadian firm compete internationally. Gomez described AI as often an equalising force within organisations because it raises the lower end of employee performance more than it benefits the top performers.
The article also notes other sovereignty‑focused projects. The Financial Times has reported on Cosine, pitched as Britain’s first fully sovereign frontier AI model: a government‑backed startup with about 30 employees and $15 million raised that has already attracted corporate partners including BAE Systems, PwC, and Lloyds Banking Group.
Gomez has warned previously that democracies’ technological, economic, and military resilience has been steadily weakening due to a growing dependence on a handful of massive technology conglomerates. He argued that while strategic resilience has a cost, the price of fragility is immense.
In conclusion, Gomez recommends that corporate leaders assess the concentration of their AI suppliers, plan for loss of access or service disruptions, and actively engineer resilient, diversified systems to secure technological sovereignty and support democratic resilience.
