Tag: artificial intelligence

  • Computer-Generated Fortune 500 cover (1965!)

    In the current era of concerns about generative AI / LLMs substituting for creative work, this anecdote about the first magazine cover to be generated by a computer (a DEC PDP-1 in this case, which apart from other hacker lore is also remembered as the hardware for the first known video game, Spacewar!) way back in 1965 reminds me that it’s still a pretty old concern that designers will transition to become prompt engineers:

    “In the course of events, Fortune’s art director, Walter Allner, might have frowned on filling the column at left with an array of abbreviations and figures, for Allner is no man to waste space on uninspired graphics. But these figures are his special brain children. They are the instructions that told a PDP-1 computer how to generate the design on this month’s cover. This program was ‘written’ to Allner’s specifications and punched into an eight-channel paper tape by Sanford Libman and John Price, whose interest in art and electronics developed at M.I.T.

    Generating the design on an oscilloscope and photographing required about three hours of computer time and occupied Price, Allner, and Libman until four one morning. Multiple exposure through two filters added color to the electron tube’s glow. […] 

    Allner confesses to certain misgivings about teaching the PDP-1 computer too much about Fortune cover design, but adds, philosophically: ‘If the computer puts art directors out of work, I’ll at least have had some on-the-job training as a design-machine programer [sic].’

    It’s not mentioned in this article, and it doesn’t look like the choices of color filters and offsets were intentional, but I have to think that this Fortune cover would look pretty amazing through a pair of 1950s-style red/blue anaglyph 3D glasses

  • Dave Karpf: On Generative AI and Satisficing

    In contrast to the flood of hyperbolic pieces about ChatGPT and other LLM-based AI (“It will revolutionize productivity!” / “It will destroy all creative fields!”), I appreciate Dave Karpf’s pointing out that these things are really best thought of as cliche generators, and that in some contexts it’s OK for the results to satisfice:

    The AI isn’t going to give you the optimal Disney World itinerary; It’s going to give you basically the same trip that everyone takes. It isn’t going to recommend the ideal recipe for your tastes; it’s just going to suggest something that works.

    And that sounds great, because both of those tasks are obnoxious time-sinks. (Yes, please, recommend a basic meal that my kids might eat! Offer me the same bog-standard Disney vacation that everyone else eventually settles on!)

  • ClipDrop: AI for Image Relighting

    This is a compelling use of AI for photographic manipulation (in my mind more practical than many of the other AI image generation examples that are flooding the art websites these days): basically the software can analyze a photograph, use AI to generate a pretty accurate depth map of the subject of the photo, and then use that for dynamic relighting (allowing you to add different artificial lights, color gels, etc.). You can try the web-based demo on your own photos! Neat.

    Demo of ClipDrop relighting an existing photo: the adjusted photo has much more colorful artificial lights applied in a realistic manner Demo of how the ClipDrop software compares to OmniData: ClipDrop's depth map is much more accurate