edenmarcotweets

gistfile1.txt

[
    {
        "id": 1767277531537744100,
        "text": "🚨 Demo Alert: risks when deploying insecure LLM agents 🤖\n(Spoiler: total environment compromise!)🚨\n\nIn a recent demo, I showcased what happens when deploying an LLM Agent to the cloud without taking basic security measures.\nThe outcome was alarming☠️\nan exploitable… https://t.co/OKoxiOUFGQ https://t.co/DPi8aVcPls",
        "time_posted": "1h ago"
    },
    {
        "id": 1766014021033972000,
        "text": "Contextual Answers from the @AI21Labs team is a fantastic model when you want the LLM to always stay within the provided context.\nThe main benefits for using this kind of model are:\n\n1.  Significant reduction in hallucinations, enhancing accuracy and reliability.\n\n2.  Improved… https://t.co/rqTuHdvOMU https://t.co/6OUIHorMrf",
        "time_posted": "2h ago"
    }
]