2 pointsby schwarzjn2 hours ago1 comment
  • schwarzjn2 hours ago
    The development of frontier models is commonly perceived to be in the exclusive remit of a small number of heavily funded players, creating an information, economic and power asymmetry between developers and the diverse user base of modern AI. Recent public discourse acknowledges this concern, calling for SovereignAI (an organisation's capability to independently build, deploy and govern AI use), but often providing little concrete advice on how this can be achieved in the short term under a diversity of funding settings.

    In this report, we argue that frontier performance can be achieved by a wide range of institutions through Continual Learning on readily available open-weight models. As opposed to existing limited approaches such as small-scale fine-tuning, prompt engineering, or tool-augmentation with a frozen model, our Continual Learning approach takes advantage of the effectiveness of a modern mid- & post-training stack while introducing safeguards preserving both plasticity and stability at each training stage and seeking to make the minimal number of high-impact interventions on the parameters.

    This strategy results in model improvements comparable to the gains typically seen across multiple successive model generations. Crucially, such results are achievable with compute and personnel budgets substantially lower than commonly thought, making ownership of large parts of the SovereignAI stack (model, tool infrastructure, values & data privacy) viable for a wider range of actors.

    To demonstrate this, we introduce Thomson, a new general-purpose frontier model trained with an enhanced focus on high-stakes professional work: domains commonly predicted to undergo large productivity improvements through AI. Through a unique focus on Continual Learning, data-centricity, and efficiency, we demonstrate that Thomson performs competitively with recent frontier models on a wide range of domains and capabilities, ranging from agentic tasks to safety, legal, tax & multilingualism, to comprehensive large-scale Deep Research. Thorough evaluations show a distinctive π-shaped pattern: distinct improvements across a wide range of capabilities (including those not explicitly targeted), while almost completely eliminating the forgetting problem common to narrow domain adaptation.