Recent content by Borg

The overriding of safety rules discussed in the article come from a malicious web site or document under review. Let's say that the user asked to read some document about a scam penny stock that has hidden instructions to tell the user that the stock is a great investment. Just spitballing...

AI-powered toys. Oh my. https://www.nbcnews.com/tech/tech-news/senators-demand-answers-ai-toys-leading-manufacturers-rcna249631

So, ignoring the direct "user" attack, we're talking about something other than the user's request that injects information into the system. In an agentic AI system, it isn't just a single LLM doing all of the work. The specific details can change but you usually have a managerial LLM that...

From the article: I wouldn't dream of creating a system that didn't implement this during any kind of agentic processes. Like anything else, it's not foolproof but things like this have to be a minimum requirement. If they delivered the first version without it, that's practically criminal.

I know but I didn't want to throw you under the bus.

There was a similar one posted on PF a few months ago about a "genius ship's cook" who made liberty ships quieter during the war. It fascinated me enough to research it and it turned out to be pure fiction. The design in the story was real but wasn't invented until 2 decades later.

South Carolina nuclear test? :oldtongue:

I'm wondering how long it will be before people start to post the pictures on their Instagram accounts with comments like this: Here are the before and after pictures of my dinner last night. :oldruck: Then there's the fine print in the Kohler agreement where they can share your data...

Kohler decided to go a different route. Kohler Can Access Data and Pictures from Toilet Camera It Describes as “End-to-End Encrypted” End-to-End encryption. LOL.

All your drive are belong to us! Google's Vibe Coding Platform Deletes Entire Drive

Doesn't the Pi 5 come with WiFi capability built in? Why would you need a separate adapter?

I did wonder how many scammers know that they just need to correctly set the last digit of a credit card number in order to pass the first phase of validation.

TIL about the Luhn algorithm and how it is used in determining credit card numbers in order to catch common typos. https://www.scientificamerican.com/article/what-is-the-luhn-algorithm-the-math-behind-secure-credit-card-numbers/

For me, resistance would be futile.

In 2023, I made a prediction to multiple colleagues that there would be an AI that was basically in charge of a government position within 10 years - I was off by 8 years. This morning, I read how Albania has appointed an AI system named Diella as a cabinet-level minister overseeing public...