@matanSF I've always got really bad vibes from Anthropic. For as much shit as OpenAI gets, to me they are just another greedy company. Anthropic however reminds me of a cult. They are dangerous.
There are 3 types of antibots companies.
The first type is the one that serves you pictures of animals asking you which one lays eggs and/or asks you to find all the squares that are part of a fire hydrant. Eventually, these type of antibot companies will start serving SAT tests as bot challenges.
The second type is the one that proudly serves you a heavily branded block page when they block you letting you know exactly who they are. They are cheap and easy to use, and allow anyone that isn't retarded to bypass them by patching puppeteer/playwright or a chromium browser for obvious leaks of automation traces.
The third type operate on the belief that they are harder to reverse because having a highly sophisticated VM, with changing seeds every 30 mins that shuffle the signals order, which are packed using their own proprietary bit packing library, with a custom base64 alphabet, and a xoring byte to further encode their UFT-8 strings will only be attempted by the most determined actor whose goals might not be money after all.
The first type of antibots will make some humans solve 20 puzzles while letting 20 other bots go through while giving antibots a bad name in general.
The second type of antibots will try to expand by taking as much market share from all the other antibots by over promising and under delivering.
The third type of antibot will sit on their laurels while allowing their main product to rotten away as their massive edge erodes away due to their inability to see beyond the present and into the future, thus they will let type 1 and type 2 antibots eat their whole fucking lunch.