Trajectory bets $300 million valuation on AI models that learn on the job

Trajectory has raised funding at a $300 million valuation to develop AI models that continuously learn and update themselves in production, an approach that could reshape AI infrastructure by automating model maintenance and reducing dependence on costly engineering teams. The startup, covered by Crypto Briefing, is betting that self-improving models will become a core layer of the AI stack as enterprises seek to cut the operational overhead of keeping systems current. The article did not disclose the funding amount, lead investors, or token details, and no on-chain metrics such as TVL or trading volumes were reported. The move matters because automated model updates could compress costs across AI-dependent businesses and shift competitive dynamics in infrastructure spending. Watch for confirmation of round size, investor names, and whether Trajectory publishes benchmarks showing its models outperforming statically trained alternatives.