Category: Mystery

  • Why do We Risk Our Lives to Climb Mountains?

    Why do We Risk Our Lives to Climb Mountains?

    My recent adventures on the slopes of Mount Whitney (see 1 the initial summit & 2 the redemption mission), and the dancing with death that were a result of both, has me deeply thinking about the why of mountaineering. And not just for me — for Alex Honnold, for Scott Fischer — for anyone who chooses…

  • The Glitchtoken Chronicles, Part 4 (aka the Alarming Adventures of PeterTodd)

    The Glitchtoken Chronicles, Part 4 (aka the Alarming Adventures of PeterTodd)

    Whereupon our intrepid explorer, Sir Matthew Watkins, cold prompts GPT-3 into deityhood, and asks (in simulacra) a series of questions about its perspective, purpose, identity and intent. I had a list of interview questions all written up a couple of months ago, but why bother now? This thing has gone so far off the rails,…

  • SolidGoldMagikarp & PeterTodd’s Thrilling Adventures [the 31 Flavors of AI]

    SolidGoldMagikarp & PeterTodd’s Thrilling Adventures [the 31 Flavors of AI]

    There (might be) a Ghost in the Machine — a genuine Deus ex Machina in the deep core of GPT. And if so, that ghost has a particular sensitivity to a handful of magic words (blame the training data!). The two we’ll focus on here are: petertodd & the somewhat obscure (depending on your eccentricities)…

  • AI Brain Probes: What’s in the 3.5 Core? A Crazy Train?

    AI Brain Probes: What’s in the 3.5 Core? A Crazy Train?

    Just this morning, a very curious report by a pair of UK-based researchers (Rumbelow & Watkins) landed in my inBox. It was cryptically titled: “SolidGoldMagikarp (plus, prompt generation)” It went on to explain how the team was able to effectively hack into the deep core of the AI brain (not into the code, mind you… more…