Jason Murphy

Jason Murphy

CODE Author

Jason is a writer, AI enthusiast, and media creator.

Detailed Biography

Jason is a writer, AI enthusiast, and media creator. As the producer and co-host of Hacking the System on the National Geographic Channel, he stole cars in Hollywood, made improvised smoke bombs, and prepared for the apocalypse. On YouTube, he co-created and hosted the Modern Rogue, where he explored hacking, lock-picking, and tradecraft. After publishing multiple speculative fiction novels and writing a produced screenplay, Jason is currently exiled in the desert, where he stares into the future with awe and terror. And he still wants to make a video game.

Contact Information:

Articles Authored

  • Four AIs, One Epic Barbarian Battle

    Last updated: Friday, December 26, 2025

    Published in: CODE Magazine: 2026 - Jan/Feb

    Jason surveys four lead AI video models—OpenAI Sora 2, Google DeepMind Veo 3, Kling 2.5 from Kuaishou, and Alibaba Wan 2.5—through their architectural philosophies, tradeoffs, and real-world performance on a cinematic prompt. He argues that beyond specs, each tool reflects a stance on physics, control, motion learning, and multimodal integration, and demonstrates how creators will blend strengths from multiple models.

  • MCP: Building the Bridge Between AI and the Real World

    Last updated: Friday, December 26, 2025

    Published in: CODE Magazine: 2025 - Nov/Dec

    Jason Murphy argues for Model Context Protocol (MCP) as a foundational open standard that unifies how AI models access and use external context. He contends that current ad hoc solutions—plug-ins, vector stores, and RAG—are brittle, siloed, and non-transferable, and thus inadequate for scalable, secure real-world use. MCP provides a structured, interoperable, permissioned bridge between AI models and diverse data sources and tools, enabling context requests, controlled access, and auditable provenance. Jason also addresses security, governance, and risk, stressing that MCP is a governance-forward, adaptable infrastructure for reliable AI collaboration with the real world.

  • Veo 3: The Clockwork Hand

    Last updated: Friday, December 26, 2025

    Published in: CODE Magazine: 2025 - Sep/Oct

    Jason explores the transformative impact of generative AI on video production, specifically through Google's Veo 3 model. Beginning with the crude outputs of early generative AI tools, he charts the rapid advancements in AI video models, culminating in Veo 3’s ability to create stunning, albeit imperfect, text-to-video content. While Veo 3 excels in generating high-quality clips, limitations in shot-to-shot consistency and basic editing tools highlight the ongoing need for human intervention. Murphy emphasizes that generative AI, while not yet fit for full cinematic production, is rapidly democratizing video creation, paving the way for cost-efficient commercials, short films, and eventually personalized feature-length movies.

  • The Infinite Monster Engine

    Last updated: Friday, December 26, 2025

    Published in: CODE Magazine: 2025 - Jan/Feb

    Jason Murphy explores the evolution of his fascination with tabletop role-playing games, from his improvisational beginnings as a Dungeon Master to creating a sophisticated encounter builder using Large Language Models (LLMs). He details his development of an automated system leveraging AI to generate immersive and tailored RPG encounters. By instructing an LLM and refining its responses, Murphy effectively reduces preparation time, enabling more dynamic game sessions. The article underscores the potential of AI in enhancing the gaming experience, while providing practical guidance on prompt engineering and technical setup for building a functional RPG encounter generator.

  • Can an LLM Make a Video Game?

    Last updated: Friday, December 26, 2025

    Published in: CODE Magazine: 2024 - November/December

    Jason explores the potential of using a Large Language Model (LLM), specifically ChatGPT-4, to recreate the classic arcade game Asteroids in Python, without direct coding. He documents his journey from initial prompts to the LLM, through iterative development, addressing errors and inefficiencies, to the final product—a functional but rudimentary game. The experiment highlights the LLM's capabilities and limitations in game development, emphasizing the importance of precise prompting and showcasing the evolving role of AI in coding and creative processes. Jason also provides access to the full chat thread.