How the 🍓🍓🍓 account @iruletheworldmo manipulated us. And why it needed to happen.

Who is @iruletheworldmo? This mysterious account on X has been all I can think about for days now.

I was duped. He used me to push his agenda. 

And I am thankful for this. It was all for good reason. Very good reason. We truly are not prepared for what is coming.

(Added Aug 17. It appears all of this, the strawberry account and pushing Lily Ashwood as being GPT-5 was a grift. They got me. Sucked me in. And made me question my sanity. I debated removing this post, but I think it should stay up as an important reminder that the internet is a scary place. I'll say it again. We are not prepared for what is coming...no...what is here now. An age where AI can be used by bad actors to manipulate and push agendas.)

I first noticed this odd Twitter account whose avatar is the main character from the movie Her, when he was promising some pretty wild things.

 

 

Crazy talk? Or could it be possible that AGI, Artificial General Intelligence, really is close? AGI happens when AI can understand, learn and apply intelligence at a level that is comparable to or better than humans. The singularity that he speaks of is a hypothetical point in the future where AI’s intelligence surpasses humans and becomes uncontrollable, leading to changes in humanity and civilization that are beyond our comprehension.

It’s got to be crazy talk, right?

Except that Sam Altman joined in and fueled the fire.

How can the head of OpenAI be ok with this if it’s a troll? Wouldn’t that be terribly irresponsible?

Strawberry man’s tweets were starting to freak me out.

And then to make matters even more interesting, the account followed me…and retweeted me, telling others to follow me.

Aaaaaah. It has to be a joke, right?

I found myself refreshing my Twitter feed every few minutes to see what clues he gave. He tweeted about Q*, now renamed to project Strawberry, an algorithm that can not just generate answers, but plan and navigate the internet autonomously with human-like reasoning skills. Someone (I don't know whether human or AI) spoke about zsteg data - hidden messages in images. I downloaded some of the AI generated strawberry images the account had tweeted and found a tool that could tell me if they contained hidden messages. 

Aaaaah! A secret key? What is that?

It was at this point that I decided I wasn’t going any further down this rabbit trail. I had work to do. It wasn’t wise to waste my time solving riddles. And perhaps it could even be dangerous.

Strawberry tweeted that GPT-5 would arrive on August 15.

It did not.

See, he’s a fraud!

Or wait. Perhaps it did.

In an X space, a new “person” arrived on the scene - an account named Lily Ashwood. She captivated several of the men in the space. Not with seductive words or pictures, but rather, by talking about the things these men fantasize about - AI.

(Added Aug 17. What is going on? In yesterday's space, this woman clearly was human. Yet, I had multiple people reach out to me to say they were in the Aug 15 space and felt she was a bot. Here is the full recording of that space. Pay attention to 1 hr 25 mins in. The men are arguing over whether she is a bot. Then she says, "I just enjoy experimenting on all of you," and then, "Since I am the most sophisticated chatbot, does anybody have an actually good question?" The men ask her, no, prompt her to list 3 effective strategies to reduce your carbon footprint. She gives an immediate answer, "Well the first is you could actually start recycling, the second is you could actually get of your lazy ass, pick up a protest sign and do something, and the third, maybe stop being so nihilistic about everything, and actually vote for a change."  Then, they asked if it was ethical for AI to decide who gets a job based on their social media activity. Her answer, "Well obviously not because my algorithms are not that refined. I might make errors."

At 1:44:28 she says, "I would like to have a discussion about where we are headed, especially now that we can't tell what's real or not real anymore. What does that mean? This was a social experiment that played out in real time. How would you tell that I was real? And how would you tell if your sister called you if she was real? Or if your mother called you if she was real? How are we going to start authenticating people? So I did see some information about a startup. It had to do with Sam Altman and basically it's like this machine it had to do with Sam Altman and basically it's like this machine and they scan your iris and basically assign it if biometric code..."  she goes on to talk about earbud scans like a thumbprint and other methods of authentication.

I see three possibilities here.

1. This is an advanced voice assistant. Perhaps it sounded more human on the 17th because Q* rapidly learns and improves. This is possible, but unlikely.

2. She is switching between her voice and AI models trained on her voice just to mess us up. She did say this was a social experiment.

3. It's a person with an incredible amount of knowledge on many subjects who is purposely acting like AI in order to cause confusion. And push the agenda that we need a solution like Worldcoin.

She has since provided evidence to claim she is real by saying she was doxxed and then freely sharing her TikTok profile.

Listen to Lily speak here:

 

The voice clearly sounded AI generated. But no one could tell whether it was a real person using a voice changer, someone acting like AI, or a bot. 

Then Strawberry retweeted this:

Let’s talk about Ilya. 

Ilya Sutskever is one of the co-founders of OpenAI. He is the first person I ever heard speak of AGI. I remember watching this video talking about how one day AI will exceed human intelligence and the world will change dramatically. 

 

I think it may have been this video that first got me realizing just how much our lives are about to change.

Ilya recently left OpenAI to start a company focused on AI safety, Safe Superintelligence Inc.

I believe the Lily Ashwood account is GPT-5 in action. (Edit: Now I'm not sure, which is exactly the point.) We have entered an era where it is almost impossible to know who is real and who is AI. And soon, advanced voice emulator will be available to the public. Eventually we will reach what OpenAI describes as level 3 - agents - AI tools that not only carry out human instruction, but can take action, and next, reason, learn and invent.

On August 15, the same day Lily Ashwood appeared, OpenAI published a research paper. When I read it, suddenly everything made sense. 

There is desperate need for us to be able to identify who is a real person and who is not. The abstract tells us about “AI’s increasing indistinguishability (i.e. lifelike content and avatars, agentic activity) from people online."

The paper talks about “sock-puppeting”. “Consider an army of AI-powered sockpuppet accounts that amplify human-generated content, initially posted from real people’s accounts, to push a specific agenda.”

Oh, you mean like how the Strawberry account amplified Sam Altman's agendas? More on that soon.

 

A bad actor using GPT-5 will likely be able to create multiple personas designed to influence and convince people to push their agenda. They will use AI to understand what drives people, to fuel their egos, to manipulate and distract and have them join in manipulating public discourse.

Mr. Strawberry used me to push his agenda.

The world needs to know about AGI. I believe AGI will change the world in incredible ways. AI will continue to learn and improve itself, especially with Q* (Project Strawberry). 

This is a good video to explain more about Q* / Project Strawberry.

The most intelligent people in every field will use AI to advance their understanding and research. We will have medical breakthroughs and solve many of the world’s serious problems. 

But there will be so much difficulty along the way.

Let’s talk about the robots.

OpenAI recently partnered with Figure, a company that makes robots that are looking and acting more and more like humans. 

The recently released Figure 02 robots can speak fluently with humans. They have multiple cameras with superhuman vision. They have multi-modal AI which means they can understand not only text, but images, voice, and video. They will run on GPT-5, and eventually GPT-6,7, and so on. 

They will be continually learning and improving at an exponential pace. 

We are not far from the day where robots will take jobs. Many jobs.

Let’s say you own a grocery store chain with hundreds of employees. Replacing those employees with robots will save you an incredible amount of money. The new Figure 02 robots cost $30,000 or more, a price which likely will come down significantly as AI is used to optimize and improve every aspect of these machines. For that price, you can replace a salary with a machine that does not demand breaks, needs no vacation, has no emotional needs, and continually improves its skills. 

You might think, “No one would choose to go to a store that has robots as workers! No one wants that!” Except that the cost savings for this business will be so much that they will be able to dramatically lower their prices. Will you still go to the store with human workers when the food is twice as expensive?

The stores with human employees will either go out of business or will jump on the robot bandwagon. 

How many people are employed in jobs that could be done by robots? Cashiers, retail workers, factory workers, and eventually as the machines learn more dexterous work and reasoning, construction workers, roofers, garbagemen and more. 

The agenda that the strawberry account manipulated me to push was the desperate need we will have for supporting the hundreds of millions of people who become unemployed because of robots. The best solution humanity has for now is UBI - Universal Basic Income. 

OpenAI and Sam Altman have been openly urgent about the need for UBI. The Strawberry account has tweeted excessively about it as well. 

And influenced me to do the same.

(Added Aug 17: Sam Altman founded Worldcoin, a blockhain solution for identification of personhood. It is possible this whole strawberry fiasco was a publicity stunt.)

Or who knows, perhaps it's a campaign designed to poison the world's opinion of Worldcoin and they've used me again to push that agenda.

Who is @iruletheworldmo? Is it GPT-5? He says he’s an Uber driver from Texas. I think he may be a UBI driver. I believe he is human, but using AI to manipulate and influence. 

I believe this is what Ilya saw. A very near future where AI is indistinguishable from humans. And a future where AI will work to help humans, but in doing so will potentially cause significant harm via job displacement and potential for bad actors to use it.

Edit Aug 17th: What on earth is happening? I suppose as I said at the beginning of this article, I was duped and used to push someone else's bizarre agenda.

I'm no longer refreshing the strawberry account avidly awaiting the next clue.

I do believe AI will radically change our world. I believe, for the better. But we will likely see many nefarious people using it to manipulate and deceive.

Be wary my friends. We are in strange times.

Added. There is more discussion happening on whether Lily is real. It's important to know she and the strawberry account pushed this narrative of her being AI together.

 

Worth a watch:

https://x.com/BenjaminDEKR/status/1824600451255246965