Howdy AI Model Mythos

OpenAI burst on the tech scene with its flagship ‘ChatGPT’ in November 2022 and new AI models have appeared since then. The tech world is fascinated and the world is not the same since then. AI chatbots and AI agents have been used to assist humans in a variety of tasks. They can write emails, do homeworks, assignments, draw images and even help you prepare for exams and quizzes and even do scams. While it seems to be helping humans do a lot of things easily, a lot of jobs have been eliminated too. Still many others fear that their job may be axed due to the rise of AI.

Programming has been vastly touched by AI. The code for my ‘Fascinating Moon Rover AI simulator’ was entirely written by ChatGPT and I was blown away. I have seen small pieces of code being written by AI, but this was quite a long program. The program was perfect in every way and I had to tweak it just a little for me to get the desired result. It is evident that human beings don’t have to write every bit of code henceforth, but they do need to roughly understand the code that is being written.

Another field that may slowly be touched is ‘cybersecurity’. While traditional concepts like defensive security, security by depth, firewalls, VPNs might remain at the core, their implementation might change drastically. As an example, a vulnerability asssessment might not have to be done the same traditional way and it can be rolled over with other roles since it can be done in a fraction of a time with new tools.

Note: Image generated by Claude AI

Here are some key points about Mythos:

1. Mythos the new AI model is dubbed as the new ”Cybersecurity reckoning” AI model
2. It is supposed to be an all powerful AI model
3. It is so powerful that it is not being released to the public as yet as it maybe misused by bad actors to cause damage
4. Traditionally, vulnerabilities are found by a structured manual approach by means of automated tools. Mythos has now discovered thousands of vulnerabilities in almost every operating system and browser that had evaded the human eye and traditional method
5. Mythos can scan large systems, identify vulnerabilities and even suggest ways that these can be exploited quickly!
6. It has discovered a 27 year old vulnerability in one of the most secure operating systems ‘OpenBSD’. It has also discovered a number of vulnerabilities in the Linux operating system that is used to run servers worldwide.
7. Considering that these vulnerabilities can be used by hackers and bad agents to exploit and bring down systems, Anthropic has decide to not release ‘Mythos’ to the public.
8. Anthropic has decided to only release it to small set of of companies like Amazon Web Services, Anthropic, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, the Linux Foundation, Microsoft, NVIDIA, and Palo Alto Networks. This is known as ‘Project Glasswing‘.
9. These companies will use ‘Mythos’ as part of their defensive work, and share their results. The vulnerabilities that are discovered in the process are patched as well.

As AI evolves and keeps us wrapped in its speed, I am curious about ‘Mythos’ and want to try it it once it is released to the public! 🙂

This post is for alphabet ‘H’ for BlogchatterA2Z by @blogchatter