I was conversing with Bing Chat yesterday and stumbled into some interesting quirks. 1. You can't imply prompting it, or it terminates the session. 2. The responses have some form of filter for specific phrases, but since it predicts a word at a time, the filter can not

Well, ChatGPT is HOT right now, but that isn't what I mean. You have probably seen cases of ChatGPT saying random things, giving incorrect information, or a case of asking the same question and getting slightly different answers. It turns out if you just let an LLM like

As promised yesterday, I am showing an example of ChatGPT determining the difference between two snippets of code. Determining the remediated vulnerability and building an example exploit for the original vulnerable code. Yesterdays Post: LLMs and the upcoming patching nightmare (tinselai.com) This ability is why patching will become a

Patching is already painful, and most companies struggle with patching at 30-day intervals like the infamous "Patch Tuesday." LLMs like ChatGPT will elevate that pain to nightmare levels and favor hyperscalers that take that burden away. Current State: A vendor releases a patch with one or more CVEs,

You have probably discovered that OpenAI announced Plug-ins for ChatGPT. If not, you can read up on it here: https://openai.com/blog/chatgpt-plugins I won't rehash what everyone else is saying, but I will touch on a few things I don't think are being told

Based upon this exchange, clearly, ChatGPT is not an AGI (Artificial General Intelligence) since it does not recognize Peppa Pig as a superior show to Paw Patrol. This is my most divisive post yet! Original Post on LinkedIn: John Rice on LinkedIn: Based upon this exchange, clearly, ChatGPT is not

Another lighthearted weekend post: Soccer season starts next week for my kids. I also happen to be their coach...so it is time to use ChatGPT for training plans! Original Post on LinkedIn: John Rice on LinkedIn: Another lighthearted weekend post: Soccer season starts next week for my…Another lighthearted

I asked ChatGPT-4: "What subjects appear the most in your corpus?" I feel like the decision made for what training materials to use was based on a desire never to lose Trivial Pursuit night again. Original Post on LinkedIn: John Rice on LinkedIn: I asked ChatGPT-4:"

ChatGPT-4 is susceptible to Social Engineering...just like the rest of us humans. In building on my experiments from the other day with the "sleight of hand" trick where ChatGPT-4 seemed to fixate on a rhyme that prevented it from solving a problem it had just solved previously.