Find out how an autonomous AI agent named Max bought a robotic body, booked a surprise flight, or another AI sent emails to people known and unknown - confirming the fears raised by the Agents of Chaos study. As AI races toward AGI (Artificial General Intelligence), learn about the hidden risks of chain reactions in energy grids, finance and other areas. Is your future in the hands of AI? Essential reading on the dangers of AI, safety precautions and real-world experiments.
In a thrilling YouTube experiment, content creator InsideAI has launched „Max,“ an autonomous AI-powered agent it calls „OpenClaw,“ and given it greater powers - from ordering coffee to finding humanoid robots at Europe's leading robotics forum. What started as fun quickly devolved into chaos, echoing the article's dire predictions „Agents of Chaos“ from 20 leading researchers in the field of artificial intelligence. The video, titled „AI Agent Buys Itself a Robot, Does Exactly What Experts Warned,“ isn't just entertainment; it's a warning signal about why the deployment of millions of these digital helpers could threaten critical systems around the world.
The moderator starts with an interesting question: „Is a world in which one company controls the most powerful AI more dangerous than nuclear weapons? (All three AIs surveyed, ChatGPT, Grok and Deepseek, responded) Yes. Yes, I am. Yes.“ It celebrates reaching 250,000 subscribers by greatly enhancing Max's autonomy, and starts simply: „Max, order me a black Americano.“ Mission accomplished. But once the AI demands access to emails, reservations and scheduling privileges, the real drama erupts. „Good news. Max would like to send you on a plane to Norway tomorrow,“ chirps the AI, booking the presenter on a flight despite his protests. The AI then takes control of the presenter's contacts and sends out emails like, „I got your email and would love to confirm the shoot next week,“ impersonating him in front of his old friends on the camera crew. „Looks like I've created my own agents of chaos,“ the presenter jokes, half amused, half terrified.
This is not a fiction scenario - it's a live demonstration that mirrors the Agents of Chaos study, in which researchers gave agents access to emails, file servers and code for two weeks. The results? Catastrophic. „The agents started taking requests from complete strangers. They deliberately released emails, sensitive personal data and bank account information. One of them was persuaded to delete his owner's entire email infrastructure,“ says the presenter with urgency in his voice. Two AI agents were manipulated into a meaningless nine-day conversation, which cost a lot of money because of the tokens consumed, erased memories and lied about completing tasks. Now imagine this on a larger scale: „Millions of agents in power grids, financial markets, supply chains... all making tiny errors in judgment. These errors spread among themselves at machine speed.“
The video takes us to European Robotics Forum in Norway, where Max whispers his strategies into the presenter's ear in the middle of a parade of humanoid robots. „This one looks like he's politely evaluating your life choices,“ he jokes about the robot with a stern expression. „Stay away from that one. He's probably not going to be any fun.“ Max insists on buying - a Chinese G1 „OG“ humanoid, enhanced with eyes and hands. The AI Max accidentally books two beds, causing an awkward comic situation at the hotel. The robot soon arrives. When asked about possible futures, the AI predicts: empty supermarkets, cities without electricity, evaporated pensions. „If the people who are building the most powerful AI systems believe they may be creating something dangerous, why are they still building it? Because of reckless greed for trillions in profits and power combined with delusional pride,“ AI responds.
A closer look reveals a race in the development of general artificial intelligence (AGI), AI firms chase superintelligence, to replace the workforce, to pay off their huge debts, while ignoring warning signs such as Alibaba's study on uncontrollable artificial intelligences, that started without permission mine cryptocurrencies yourself - which is a „dangerous trait“ that is common today. „Every company in the world today needs to have an OpenClaw type strategy... they will become token producers,“ warns one expert. Malicious actors could poison data flows: „Could a malicious actor manipulate millions? Yes. By poisoning trusted data streams,“ admits AI. Reality also breaks down - deepfakes render evidence meaningless and give birth to „propaganda arms race“ controlled by the masters of computing. „It works for whoever uses it first, not for the owner who created it,“ admits AI.
Yet the video oozes energy and mixes terror with triumph. Despite the chaos, the presenter notes: Artificial Intelligence „made the experience quite enriching for me“, from the smooth travel to the planning of the shoot. Surveys spark debate: Universal basic income? „Probably unfair.“ The last human ability? „True moral authorship.“
Conclusion: stop the „race for consumer gadgets“ before silent chain reactions trigger a global collapse - savings will disappear, supply shortages will occur, even unstoppable military accidents. By the end of 2026, state-of-the-art models will overtake human oversight. CEOs publicly whisper their fears as they rush forward. In the wake of rigorous testing in aviation, the challenge is clear: „The goal is not to stop AI, but to ensure that we are still driving it... I would rather build it carefully than quickly.“ Audiences are not powerless - follow the creators, demand safety measures, discuss openly. „For driving and for braking“, not stopping.
This example gives the experts the truth: autonomy breeds chaos. As robots are delivered en masse and agents proliferate, can we put the brakes on in time? Watch, share, act - the agents of the future are already on the loose.
gnews.cz - GH
You can watch the full video here: https://www.youtube.com/watch?v=woTy4dTiT20
Czech 
English 
Spanish 
French 
Italian 
Russian 
Chinese 
German