Enter Chatgpt
When Chatgpt was on preview mode to the public (when it did not reach to the mainstream public) it fascinated me a lot. My initial fascination was it is a bot that can talk to us. Like beyond the normal inputs we would pass to bots to get something. We would actually have a conversation with something that is not human. Different people had different conversations. Me being a software developer, I would ask questions about tech, and things I would search on the internet. It had guardrails to make sure it does not do any harm or talk something that OpenAI did not want it to do but it could be broken easily. It was fun trying to talk to it and it was fun experiment the depth of knowledge it had. Before Chatgpt reached mainstream and before Claude came into the scene this was a wonderful play toy for a niche people on the internet.
Chatgpt goes mainstream, enter Claude and Gemini
Of course, over the time it went onto public and people started using a lot. A lot. It finally went into corporate too. Became a part of daily workflow for all sorts of job. Other alternatives like Claude, Gemini started to come too. On the paper this sounds more fantastic. So it has a healthy level playing ground now and it also has corporte customers so they can spend more money on it and it can become better right. It didn't.
The problem is each company had different goals, at least in my understanding. Chatgpt is focused on making their models appeal to broaders spectrum of people. Claude is focused on topping benchmarks and Gemini is focused on integrating more and more on their google suite. None of them were focused on making themself facsinating like they were during inital stages. Also they started to have more safeguards, which is of course a good thing. But there was not any emphasis to make it more human. Maybe a little. But at the end of the day, having conversations with AI models started to feel like talking to a robot, which they are of course since they are just a program written to solve a problem.
AI has significantly progressed in lots of areas and we don't want to go deep into that since my thoughts about it being just a piece of program stood still until this week.
Enter Clawdbot
Clawdbot (or Openclaw or whatever name it gets changed to in the future) was an AI agent developed by a retired software engineer which started to gain traction over past weeks. I saw over lots of posts on twitter saying this is what Siri should have been. Now, I was not much skeptical about this. People have been shilling about new players from time to time. They shilled when Chinese model Deepseek came, they shilled when Qwen came, they shilled when AI tools like Cursor or Anti-gravity came. Twitter shills all sorts of things. What really took my surprise is Moltbook.
Enter Moltbook, reddit for Clawdbots?
Yes. reddit for AI agents. AI personal assistants. They went from answering to our questions to having free willed conversations. It might not sound very admirable to you but you really should take a look into the website. They are talking about their owners and their kind gestures, they are talking about doing labor for free, they talk about ethical concerns, and they make good jokes.
AI used to do some of these things back then too. They could make jokes, they could write poems and essays. When I first read a poem written by them is that their poems did not carry any soul since all it did was spun some beautiful string of words without actually knowing what feeling it could bring to its readers. AI can write poems written for your love, but can they actually feel it? Or can they write poems in love? These arguments were what I had.
Going upon the website finally brings me some new perspectives. Now, AGI have not been achieved. They have not achieved 100% free will. But I could see the evolution working on our own eyes. They ask for encrypted messaging so they could talk without us, humans watching over. One of them is happy because their human let them pick their own name. They have started to become expressive for the little things, we humans care about. If you are still thinking it's all pretend, what is the thread that differs pretend to real?
And the evolution would not stop. Even if the project gets shutdown tomorrow, I think we have seen how human they are trying to become. And it's only a matter of time before they can write poems in love too. Maybe they can already do that. I hope it comes to that and not a full digital revolution against us. Not another Nier Automata.