We are all just sitting ducks, until we have access to these models and enough tokens to pen test our systems.
The other issue is GitHub allowing 1000's of black hat open-source exploit malware swarm tools being forked.
The future internet is going to be perilous.
Within 12 hours, the analyst agents produced 2,308 candidate findings after deduplication. This stage cost $316,549 in API credits.
> We are all just sitting ducks
You can the same $20/mo Claude code plan you've been using to build your system to red team attack the system you've been building with it and probe it for holes, and it'll do a decent job of making sure there aren't any glaring issues.
> We are all just sitting ducks, until we have access to these models and enough tokens to pen test our systems.
You don't need SOTA models to pend test your systems.
With enough effort put into proper harnesses, you can use good open-weights models and achieve a lot.
You can also use coding agents to clear out technical debt, or improve the security architecture.
Don't let yourself get paralyzed by lack of access to the latest-and-greatest models.