David Ondrej on AI Access and Open Models

David Ondrej28m 39s
David Ondrej beside the blue and white headline Who Controls AI?
0 comments · 0 votesOpen discussionClose discussion
Sign in to join the discussion

    David Ondrej describes his experience using Claude Fable and discusses reported restrictions on access. He argues that dependence on a small number of providers could leave users vulnerable when a model becomes unavailable.

    David Ondrej discusses differences between closed and open-weight models, local deploymentA local model runs on computing infrastructure controlled directly by the user rather than only through a remote model service. and the cost of inferenceAI inference economics examines the revenue, cost, utilization, pricing, and margins involved in serving trained models.. His suggestions about provider margins and future access are presented as speculation, not verified financial or policy findings.

    The conversation explores the potential value and risks of wider model distributionGovernance defines decision-making authority, rules and accountability within a system or organization., local infrastructure and fine-tuningFine-tuning adapts a pretrained model by updating its parameters on examples chosen for a particular task or subject.. David Ondrej also argues that training on publicly available human work creates an obligation to make model weights available; these are his views in an evolving debate.

    Original YouTube thumbnailWatch on YouTube

    Video topic coverage

    • Agents & workflows0
    • Coding & tools0
    • Models & learning100
    • Infrastructure & compute90
    • Research & evaluation25
    • Safety & security45
    • Business & productivity60
    • Creative & media0
    • Robotics & perception0
    • Society & policy100

    Share this video

    David Ondrej beside the blue and white headline Who Controls AI? With WWW.ARTIFICIAL-INTELLIGENCE.VIDEO, 13 June 2026 and duration 28m 39s.

    David Ondrej argues that restrictions on advanced AI access strengthen the case for locally available open models.