Project Glasswing - Anthropic's next big thing

If you have not already seen it, you should read some of the conversations surrounding Anthropic's newest LLM, Mythos. It scared me reading through the official releases. If AI is able to crack through vulnerabilities that have been missed for years and years, how are we meant to defend ourselves? Does everyone need to bend the knee to the AI companies and beg, "Help us"?

As usual, with these LLM/AI companies, it is tough to decide if this is hype or truth. The only telling sign that I noticed is why would some of the other tech giants want to partner themselves with Anthropic, unless they had some secret sauce?

Genuinely, I like Anthropic. I have liked them since they stated their morals and turned down the Pentagon. That is not to say I agree with every decision, such as the rate limits that seem to be shrinking, but I believe there is something a little more... human? with them.

I hope that Anthropic is trying to better the systems that we need to be secure and is not only asking to be let into other's hidden codebases. Maybe they will steal everyone's code and perform the Great Code Heist of 2026?

Only time will tell. As usual.

Disinformation in the Age of Information

I have been plagued by how hard it is to truly come to a conclusion on the internet now. Sources buried or never found; reddit comments cited as truth in LLMS. Do we need to go back to pulling out encyclopedias at the local library and getting buried in the stacks? Are books even relevant anymore?

With how powerful agents and bots are at writing cohesive thoughts, is any forum sacred any more? I may have mentioned this in a previous post, but I hope AI tagging becomes a thing, where any bot/agent writing posts must claim that it is AI. But how do you enforce that? Bot detection will never be good enough (someone prove me wrong?).

Cadence of posts moving forward

No clue. 😊