7 ms·

This is amazing and terrifying (I am a security engineer and parsing complex document formats is a never-ending treasure trove of vulnerabilities).

• tashian 2y ago
  AI agents run in isolated VMs, but PDFs have been out here running in the open for 30 years!
  • miohtama 2y ago
    But can your PDF run an AI agent?
    • freedomben 2y ago
      Looking forward to a day when you may not have a powerful enough GPU to open a PDF
    • hnlmorg 2y ago
      In my opinion the question isn’t so much “if” but rather “when”. When will AI research and hardware capabilities reach a point that it’s practical to embed something like that into a regular document? We’ve already seen proof of concept LLMs embedded into OpenType fonts. I guess the other question is then “what capabilities would these AI agents have?” You’d hope just permission to present within that document. But that depends entirely on what unpatched vulnerabilities are lurking (such as the Microsoft ANSI RCE also featured on the HN front page)
      • btown 2y ago
        For Chrome's PDF renderer, the runtime is V8, so we're literally one (hilarious) line of code away from this glorious future existing today: https://pdfium.googlesource.com/pdfium/+/refs/heads/main/fpdfsdk/fpdf_view.cpp#1210 https://pdfium.googlesource.com/pdfium/+/refs/heads/main/fpd... > // Use interpreted JS only to avoid RWX pages in our address space. Also, --jitless implies --no-expose-wasm, which reduce exposure since no PDF should contain web assembly. > return "--jitless";
        • Thorrez 2y ago
          You could write an LLM in plain JS, right?
          • btown 2y ago
            Yep, but one without the ability to even JIT down to vectorized CPU commands (to say nothing of GPU connectivity) would be incredibly slow indeed!
    • siva7 2y ago
      The first widespread AI Malware will be a historic moment in this century. It will adapt like a real biological virus to its host and we have no cure for this.
      • saagarjha 2y ago
        We could unplug all the GPUs.
    • Swizec 2y ago
      > But can your PDF run an AI agent? Oh it's so much worse than that. Your font can run an AI agent. Llama.ttf: A font which is also an LLM -- https://news.ycombinator.com/item?id=40766791 https://news.ycombinator.com/item?id=40766791
      • bawolff 2y ago
        Well a font using a custom experimental shaping library. Your font can't do it normally.
      • belowm 2y ago
        Crazy. Looking forward shipping apps as .ttf instead of docker images.
      • erk__ 2y ago
        You can also play Tetris in a font: https://www.youtube.com/watch?v=Ms1Drb9Vw9M&t=1370s https://www.youtube.com/watch?v=Ms1Drb9Vw9M&t=1370s (disclaimer: own work)
• neuroelectron 2y ago
  This isn't even the beginning of what's possible in PDFs.
• wayvey 2y ago
  The amount of attack surface in various format parsers is pretty stunning and terrifying indeed
• mizzao 2y ago
  The "code execution" in PDF parsing is what enabled this legendary zero-click, zero-day exploit of iOS devices: https://googleprojectzero.blogspot.com/2021/12/a-deep-dive-into-nso-zero-click.html https://googleprojectzero.blogspot.com/2021/12/a-deep-dive-i...
  • kccqzy 2y ago
    That exploit is indeed legendary but the code execution involved is not JavaScript. In fact the iOS PDF renderer does not have JavaScript enabled.
    • saagarjha 2y ago
      Obviously a skill issue; a true hacker would re-enable it.
• enews01 2y ago
  Theres a malaysian movie where the main premise is a hacker who uses pdf executions to steal one cent from every persons bank account. Its pretty interesting.
  • brettermeier 2y ago
    Do you know the name of the movie?
    • wastholm 2y ago
      Not OP, but I found a series, not a movie, titled _One Cent Thief_ that fits the description. Sounds interesting. https://archive.org/details/OneCentThiefSeries https://archive.org/details/OneCentThiefSeries