Bruse Schneier has an excellent article on how large language model AIs are being maliciously manipulated because of an old-style vulnerability being exploited. AT&T was probably the first organisation hit by such an attack that exploits systems that use the same channel for both data and commands. Back in the …
LLM AI Vulnerabilities
Here is an excellent essay by Baldur Bjarnason on the manipulation of ChatGPT and other large language model AIs: The poisoning of ChatGPT (softwarecrisis.dev)
Listen to the programme written by ChatGPT
BBC Click presenter Lara Lewington set out to create a radio programme using ChatGPT. Why not listen to it whilst you have your lunch? A Documentary: By ChatGPT – BBC Sounds It is really worth a listen, to get a good grasp as to what ChatGPT is, what it can …
Continue reading “Listen to the programme written by ChatGPT”
What would tech writers write about if ChatGPT did not exist?
ChatGPT has been used for homework, writing radio programmes, academic papers, articles, coding, hacking etc. etc. etc.. Here are some of the most recent articles examining privacy and cyber crime using artificial intelligence, of which the adapting, convincing email scams is probably the issue that organisations need to think about …
Continue reading “What would tech writers write about if ChatGPT did not exist?”
Microsoft 365 and ChatGPT
I gave a lunchtime online talk this week to a group business professionals, answering ten set questions that ranged from “did I think robots would take over the world” to “what were my top tips for a small business with limited finance when it came to cyber security”. AI and …
Poisoning AI output
We all know (unless you have had your head under a pillow) that ChatGPT is an artificial intelligence system that among its capabilities it can write a love poem, and academic paper, computer code or even your homework. It can also answer your questions – just ask Bing! ChatGPT AI …
